Back to LSB Blog

Social Media Testing: 5 Steps to Test Your Way to Success

Do you know if your content is connecting with your consumer? Are there creative variations you could make to improve engagement? You may be surprised at the insights you can uncover when performing a social media test campaign for your brand.

Social media testing is a topic marketing folks talk about a lot. Did it work? How do we know? The conversation typically includes gut feelings and engagement numbers but no concrete answers on how the post performed for the target or, better yet, could it have performed even better?

Where to begin?

It’s important to thoughtfully select the social media platform to host your test. Facebook is a great place to start for several reasons.

  • Largest audience of any social network with 1.23 billion daily active users.
  • Most robust reporting of any social network.
  • Highly sophisticated targeting capabilities including core audiences, custom audiences, and look-a-like audiences.

What if Instagram is your sweet spot? Good news. Facebook owns Instagram so you can take advantage of all the features listed above. However, the ad delivery ratio on Instagram compared to Facebook is 1 to 5. That means your ads will be shown less, since there is less inventory. Will your sample size be large enough to draw real conclusions on Instagram? It’s something to consider.

If your brand isn’t active on Facebook or Instagram (seriously? ok), social media testing can also take place on other platforms such as Twitter or LinkedIn. Follow the general steps below and you will still walk away with insights to help guide your content.

1. Make a social media testing plan.

In social media testing, planning and set up is KEY. Understand what questions you want to have answered and what variables you can test to find those answers. It’s also imperative you are very clear and focused with your choices. There are an unlimited number of variables in social media testing but too many can muddy the results. When testing content, you must do your best to isolate the variables you’re testing.

Sample questions to get you started:

  • What type of creative is most engaging? What is our key metric for engagement?
  • What tone of voice engages users most?
  • Which creative visuals are driving the most clicks?
  • What content drives the most conversions? What conversion metric is most important? Only pick one.
  • How do we measure success? Clicks? Likes, comments, shares? Link clicks?

Once you understand the questions, you can start to identify the variables you will need for a productive test.

Know Your Variables:

  • Target variables – Age, location
  • Creative format – Static image, GIF, video, 360 video, carousel
  • Creative placement – Facebook newsfeed, Facebook right-hand side, Instagram, Mobile, Desktop
  • Creative design – Infographic, lifestyle, product-focused
  • Creative tone – select three unique copy styles

Once variables are defined, you’ll need to understand where and how you are going to test those variables. Some variables you’ll want to implement in ad set-up. Others are easily pulled on the backend through Facebook’s ads manager. Carefully plot out where you will access those insights.

Here is a good place to start:

Tools for Tracking Success:

  • Facebook Ads Manager – Performance and Breakdown (Image below)
  • Facebook Pixels – Set up Facebook pixels to track conversions on site
  • Google Analytics – Set up UTM codes to track all ads driving to site

Social Media Testing

2. Understand your target.

The depth of targeting capabilities can be overwhelming on Facebook. Bring it back to the questions you want to have answered. Do you want to target current customers or potential customers, geographic locations, or specific age groups?

In your first test, go as broad as you can to ensure you are testing some of your own biases. You might just be surprised.

When running a Whole Foods coupon promotion for a specialty CPG client, we opened up our targeting to both men and women and found a majority of the Facebook ad engagement was coming from men. This went against some of our existing research and presented an insight that encouraged further testing.

Up for a challenge? Try Facebook’s split testing. This feature allows you test the same content across different audiences and performance variables. Check out this step by step guide from Jon Loomer on how to set up split testing for more info.

3. Content creation.

Finally, the fun part. To start, pick one piece of content. Whether it’s a whitepaper, a coupon landing page, or even a news story showcasing one of your Hospital’s service lines. This is also the time and the place to test new creative executions – try 360 video, GIFs, or carousel ads.

In the example above, the Forbes article is the same in both posts. The landing page is the same. The message is the same. This creates a more accurate test knowing consumers are being shown similar content using different creative executions to understand performance.

We recommend running these social media tests as dark posts on Facebook. Dark posts will add an additional layer of quality assurance. If posted organically, several different uncontrollable variables come into play (time posted, which was posted first, organic engagement etc).

Once you know the winning execution from dark post testing, you can improve the content you post organically on your Facebook page.

A brand’s content strategy should be ever-evolving. Social and digital media trends change too quickly to remain complacent. A core part of our social media team at LSB is our in-house content studio, allowing us to test and then iterate content quickly and efficiently.

An example of an A/B copy test is below. When set up in this manner you will understand the top performing Tone of Voice when completed. We recommend including two to three unique tones of voice.

 

Social Media testing

Similarly, you also want to A/B test creative. Use a minimum of three different creative executions.

To do this effectively, a copy test and a creative test will need to run as two separate ad sets. Which leads us nicely into…

4. Organization and launch.

Time to put all your hard work into action.

 Organize all your assets in one place. You now should have the following:

  • Question(s) you want to have answered
  • Variables selected with corresponding metrics
  • Copy variations (2-3)
  • Creative assets (3-5)
  • Tracking measures in place – UTM codes, Facebook pixels

Set up two different tests to isolate both copy and creative variables. This will include multiple ad sets and ads within those. A brief example is below.

When setting up Facebook ads, make sure your naming conventions are consistent across the board. Include the variables you’re testing in the name. This will make life exponentially easier when reporting rolls around.

Check and double check target audience, ad placements, creative and all other variables are correctly implemented.

Launch!

5. Reporting.

Because of your planning and organization, reporting should be a cinch.

The data deluge is real. Stick to the questions you laid out to answer at the onset. Use the metrics you identified as being the most critical to your brand. Report concrete and actionable findings and use them to improve your content.

For additional steps to ensure your reporting is top notch, take a couple lessons from our very own Todd LaBeau and his five step plan for successful campaign measurement.

Through social media testing and thorough reporting we’ve uncovered which age groups engage most with different types of content, confirmed a brand’s core consumer demographic, determined tone of voice to be implemented across brand communications and more.

The more you test the more you learn. Happy testing.

 

Interested in having the pros do social media testing for your brand? Let’s see what we can discover. Contact LSB today.

More...

Categories:

Analytics, Content Studio, Digital, Social Media

Subscribe to Newsletter

Get the latest news and information direct from us to you!
CONTACT US - info@lsb.com