Why Split Testing Is the Best Way to Prevent Wasting Marketing Money
In their book , business coach and consultant Dan S. Kennedy and marketing strategist Kim Walsh-Phillips show you how to use direct response marketing principles on a variety of social media platforms to drive real results and profit. In this edited excerpt, the authors describe why you should stop assuming and start split testing your ads.
I have no desire to throw away a single dollar of my hard-earned money. And I am sure the same is true for you. So another good reason why social media should be a tool in any sophisticated marketer’s arsenal is the ability to test theories and optimize accordingly in a moment’s notice.
Unlike direct mail, print advertising, or other media channels, there's no need to invest significant dollars or wait 30, 60, or 90 days to see if version A, B, or C of your ad works. Tracked correctly, in a day or two, you'll know exactly what the results of your different ads are and you can scale up or down accordingly.
Great marketing that produces monetary results is based on facts, not guesses. Optimized results are developed by examining data and changing campaigns accordingly.
In comes the power of social media. In real time, you can optimize an ad campaign based on an audience’s behaviors, likes, habits, and hobbies. You can instantly scale advertising spend up or down depending on the results you're getting.
The reality is though, that very few businesses use even 20 percent of social media’s power. So they're wasting their dollars again and again in this channel. They guess instead of test. They assume instead of research.
One rule of marketing that I've found to be true is that assumptions are usually wrong. When I'm speaking to an audience or presenting online training, I share some of our top split tests. And 90 percent of the time, the audience gets it wrong.
That's right: Nine times out of 10, they're wrong.
Meaning, that if ads are only created on assumptions as to what will work the best, then ads would never be optimized. Just think how much money is wasted! Just think of all the vacations, at the beach, with a hot sun, a cool breeze, and an ice cold drink in your hand, that could have been taken with this money. Stop assuming and start testing.
In my career, I'm blessed to be able to do a lot of professional speaking and am often asked questions about the best ways to execute strategies. I get questions on how often to post a new Facebook status, the best times to send out an email, or the most effective headlines to generate the biggest response. While I can offer general best practices (post at most once per day, don’t send out your emails on a Monday, and make the headline about the target audience’s pain), I can’t offer specific strategies without one very important step — testing. Testing communications strategies to a small segment of the target audience before launching to the entire audience is a way to gather data on the most effective method of communication. Testing small is the number one way we prevent wasting money in our clients’ marketing campaigns and can show continuous improvement.
Imagine an ice cream manufacturer releasing a new flavor. It doesn’t just put something together and send it directly to the stores for sales. It creates several versions, tests in-house, tests out-of-house, and then chooses the option that performs best.
Your marketing should be treated the same way — with no guesses or assumptions. Take the emotion out of your marketing, and execute your strategies based on measurable results from your testing.
When testing, keep the following social media ad-testing best practices in mind:
- Track conversions, not clicks. You want to optimize your overall ROI, not waste dollars on clicks. An ad that gets 100 clicks is not better than an ad than only gets 25, if the second ad gets more conversions.
- Set up the target market demographics so you're always testing only one thing at a time. If you combine too many variables in your ads, it will be difficult to optimize them well.
- Start with split testing the demographic, then use the best performing demographic in the rest of the tests.
- For the ad image that performs the best, test changing the ad’s coloring to see if it performs better. For instance, does a black background perform better than blue or white? Better than red?
- Take the best coloring of the ad, and test different wording to see if that makes a difference. For example, if you tell them to “click here” vs. “download now,” does that do better?
- Test adding a call-to-action button vs. running without a button. For instance, does adding a call to action hurt sales because your social media post now looks like an ad or does it help sales because there's a clear path the prospect can take for action?
When you've run all your optimization tests, it's time to move onto the landing page.
Here’s a quick checklist of the items to test for:
- Ad images
- Target market demographics, interests, and behaviors
- Landing page copy
- Opt-in form
- Text below the photo
- When the ad is run (time of day, day of week, week of month, month of year)
- Daily budget
- Type of bid, automated or manual
- Placement (newsfeed, right side, mobile app, mobile and/or desktop)
Unfortunately, social media is like any other form of marketing: If you want a continued high ROI, you have to keep testing.