Unlike e-mail or mobile marketing that have a one-way delivery channel, social media can distribute your messages on sites like Twitter, Facebook, YouTube, Pinterest and many others. You can effectively monitor the conversation surrounding your business and launch your message when the timing is best.
While it provides numerous benefits in terms of release and overall reach, social media also raises some challenges in terms of determining how and when it’s best to engage your audience.
Testing everything from which social media sites are more appropriate for your business, to timing, type of content and whether or not to insert an offer or keep it conversational, can make or break your marketing efforts.
Here is a breakdown of the key pointers of split testing in social media:
Create A Brand Persona
One of the main factors in successfully marketing your business is knowing who your audience is and how to effectively reach them. Ask yourself these questions:
- How do your customers prefer to digest content? E.g. Newsletters, online magazines etc.
- How do they stumble upon, digest and share content?
- What are their main search topics and what do they talk about on on the social web?
The answers to these key questions will shed a light on the content of your web media, the kind of social media that will best match your brand profile and strategies to get more advocates to various channels (e.g. your Facebook Business Page, Twitter profile, blog or company site).
Rapleaf is a great online tracking service that helps you understand your audience and predict their actions. You just send them a list of your users and their email addresses and they provide information about the person using that email address (e.g. age, gender, location etc.). Their service is totally free.
In a nutshell, split testing allows you to test with two variables (A and B) and is probably the easiest to observe. Let’s consider Facebook ads as an example. You have Ad A and Ad B – both have four key variables: headline, copy, photo and bid amount.
The first rule is to only change one variable at a time: you can keep the same headline, copy and bid for one ad and test two different types of images. The results will determine which one works better for your ads.
If possible, run the ad under similar conditions: same time of the day, same bid amount, same duration, etc. Never try to tweak an old ad – you can easily create a new ad using the Facebook feature “Create a Similar Ad”.
Here are three variables to consider for split testing:
1. Social media buttons
Don’t be afraid to remove the social media buttons from your website. Remember, readers don’t come out of nowhere – they don’t land on your website and then just run to the social sites to promote you only because you have social buttons on your site. Users discover content via Facebook, Twitter, Google+, Pinterest etc., not the other way around.
Try and remove Facebook buttons and see how traffic from Facebook may increase. You may be surprised! The reason? Instead of “Liking” your content, users will share it on their timeline. If you provide great content, people will take the time to read and then copy & paste your URL it in their networks.
Running social media campaigns is not a 9-5 job. Use Followerwonk to determine when most of your audience is actually active in social media. Use the information to plan your daily and weekly social calendar accordingly.
Most companies either share too much or too little content. If you’re sharing your own content a few times a day, try and scale back how much you are posting your content and use that time to actively engage with users (e.g. answering questions, retweeting interesting Tweets, posting comments, etc).
Tools To Start Boosting Your Social Media Campaigns:
- Use CrazyEgg to find the most-viewed areas on your site (a heatmap – see below)
As you continue to experiment with different variables, you will get a better understanding of what works and what doesn’t, which areas you need to improve and what offers your customers are most responsive to.
When you’ve identified the pros and the cons of your initial strategy, you can organize your final marketing campaign knowing there is nothing left unchecked in terms of research and testing.