Editorial: Set Yourself Up to Measure Your Own Social Media Questions

Seth ReslerBy: Seth Resler, Jacobs Media

I often get asked very specific questions about social media, such as:

  • “How often should we post?”
  • “What topics should we post about?”
  • “Should we post videos directly to Facebook or use a YouTube link?”

Inevitably, I give the same answer: “Experiment and see.” I don’t say this as a cop-out, but because what works for other stations (or other companies) may not work for yours. Yes, there are suggested best practices out there, but you should never let those take the place of hard data.

As broadcasters, this is a relatively new idea for us. We often program by our “gut.” A record or a contest or a bit either sounds good on the air or it doesn’t; Nielsen doesn’t give us data that’s granular enough for us to pinpoint specific results, so we offer our best educated guess.

But online, we have much better analytics. We can run small experiments and see the results in real time. We don’t have to guess which is driving more traffic to our website, Facebook, or Twitter; we can actually see the answer.

A great example of how to gather this data comes from the folks at NPR. They had questions about the best time to publish Facebook posts. They had a theory — called the “Facebook Whale” — and they set out to test it. Bryan Wright and Lori Todd explain what they did and what they learned here.

How can your radio station perform similar experiments to answer your social media questions? Follow these steps:

1. Define and agree upon your metrics.

Start by asking, “What does success look like?” What is the goal of your social media efforts? To get lots of likes? To drive traffic to your station’s website? To add registrants to your email database? To increase ratings? To generate revenue? These things are all related to each other, but some are more important than the others. Make sure that everybody agrees ahead of time on what the appropriate unit of measurement is and how many constitute success.

For example, let’s say your station has its annual Spring Fling Concert coming up and you’re thinking of running Facebook ads to help sell the show. You need to fill 1,000 seats to break even and 2,000 seats to hit your revenue projections. If you sell more than 2,000 tickets, your boss will love you; if you sell less than 1,000, you’ll need to update your resumé.

At the end of your experiment, you don’t want different people to look at the same result and draw different conclusions. If you’re measuring website clicks but your General Manager only cares about ticket sales, you’re going to run into problems. By agreeing upon the proper metrics ahead of time, you can avoid this confusion. You and your GM agree that while you should monitor website clicks, success will ultimately be measured by the number of tickets sold.

2. Set yourself up to measure.

Now that you’ve decided what you’re going to measure, you need to make sure that you can measure it. Make sure that you have the appropriate tools in place, whether it’s Google Analytics, Facebook Insights, Bit.ly reports, etc. Also, make sure that you understand how to use these tools. If you don’t have them or don’t understand them, you’ll need to address these issues before you run any experiments.

In the case of our Spring Fling Concert, we can set up different website landing pages, which are identical except for a hidden code passed to the ticketing service. This code allows you to pull a report to see how many of our ticket sales came through Landing Page A and how many came through Landing Page B.

3. Run experiments.

One of the simplest experiments to run is called an “A/B Test.” Control (to the best of your ability) all of the possible variables except one. Change that one variable for half of the test and measure the results.

For example, you could create two Facebook ads that are identical except for the headline: One reads, “Tickets to the WKRP Spring Fling Concert are on sale now” and links to Landing Page A while the other says, “See who’s playing the WKRP Spring Fling Concert” and drives people to Landing Page B. Set the Facebook ads to alternate so they are both shown an equal number of times. Watch to see which ad produces the most website clicks and, more importantly, results in the most ticket sales.

4. Review the results together.

Be sure to set aside time to review the results of your experiment with your team. Discuss the results and draw conclusions together. This ensures that everybody is on the same page.

After a week, let’s see which of the two Facebook ads has produced the most ticket sales. Interestingly, Headline B (“See who’s playing…”) resulted in more clicks, but fewer sales than Headline A (“Tickets…are on sale now”). While Headline B was attracting people who wanted to see the lineup, they apparently aren’t ready to make a purchase. As a group, you and your GM may decide that your budget is better spent on Headline A and shift your dollars accordingly.

The best way to figure out the proper digital strategy for your radio station is to set yourself to perform small experiments like this. If you would like help doing so, feel free to reach out to me.

Editor’s Note: The views and opinions of the above article do not necessarily reflect those of the MAB. Contact the MAB for information on the MAB’s official editorial policy.

Leave a Reply

Your email address will not be published. Required fields are marked *