SCIENCE

Having trouble deciding if it’s worth running a paid Facebook campaign? It’s a difficult question and a hot topic as Facebook continues to force companies to pay for engagement. Adweek reported that Facebook will continue to slash brand’s organic reach down between 1% to 2% in the upcoming future. That means for every 1,000 followers only 10-20 of them will see your page’s post. This is a blow to a lot of companies’ egos because they spent years building up active communities.

At least for the short term it appears the only way your business is going to reach followers and engage with new audiences is to pay for it. You need to ask yourself, is it worth it to my business to run paid Facebook campaigns? Will Facebook deliver the results I’m looking for?

A good place to place to start answering these questions is by employing the scientific method.

The scientific method is a set of techniques designed for acquiring new knowledge. Here at Inflow, we regularly use it to investigate and test hypotheses to better achieve and deliver results for our clients. Here’s how to use it for your next paid Facebook campaign.

Steps of the Scientific Method

Remember back in 6th grade when you learned the scientific method to test Newton’s Laws of Motion? No? Fantastic, let’s do a quick review:

  1. State the Problem
  2. Make Observations (Research).
  3. Construct a Hypothesis.
  4. Design and Perform an Experiment to Test the Hypothesis.
  5. Analyze Your Data to Determine Whether to Accept or Reject the Hypothesis.
  6. Bonus: Propose and Test a New Hypothesis.

1. State the Problem

Every business has problems and they’re typically simple to identify. It’s solving the problem that is often difficult.

  • Define the problem (e.g. we need more white paper downloads)
  • Now state the problem in the form of a question
  • Example: How can we increase our white paper downloads?

2. Make Observations

Our team starts by answering some basic observational questions looking for potential opportunities. Use these questions to stimulate your thinking:

  • What have we tried before (think previous hypotheses)? What worked and what didn’t?
  • What am I doing now to drive traffic?
  • What type of content are we publishing?
  • Who is my target audience?
  • Where is my target audience? Online? Offline?
  • Is there a match between my content and users on certain platforms?

Answering these questions provides a much better understanding of your client’s audience. This type of research and observation will inform a lot of decisions moving forward so make sure you’re thorough.

  • Example: Our target audience spends a lot of time, and is very active, on Facebook but not Twitter.

3. Propose a Hypothesis

Based on the observations made, it’s now time to formulate a hypothesis, which is a statement that can be used to predict the outcome of future observations. Remember to make sure that your hypothesis is a specific statement relating to a single experiment.

  • Use the classic If/Then statement to make an educated guess.
  • Think: “If _______(I do this), Then _______(this) will happen.
  • Example: If we use Facebook to promote our new white paper, then we will increase downloads.

4. Design and Perform an Experiment to Test the Hypothesis

Now is the time to see how the real world reacts to your hypothesis. This is the Who, What, Why, When, Where, How, and for How Long of your experiment.

  • How will you set up your experiment?
  • What tools do you need?
  • Are there any prerequisites (access to accounts, creating new accounts)?
  • What kind of budget?
  • How will you measure success?
  • How long will the campaign run?
  • When are we going to run the campaign?

5. Analyze Your Data to Determine Whether to Accept or Reject the Hypothesis

Start by answering these questions:

  • Did the experiment support your hypothesis? Why or why not?
  • Based on the goals of your experiment, did Facebook produce results?
  • Is it working at the pace you thought it would?
  • Would you recommend continuing to use this marketing tactic and/or platform?
  • BONUS: What could be done to improve the results?

Keep in mind that while it’s good to give a new marketing tactic time to work, you won’t necessarily need to keep doing something that is not working (e.g. wasting time and money) purely for the sake of the experiment. That is where marketing can be a little more fluid than scientific research.

Some Takeaways…

The goal of scientific method is to acquire new information on how things work. It’s important to remember that it is designed to minimize the influence of the experimenter’s bias on the outcome. You may have a preference for a desired outcome when you start your hypothesis, but it’s important that your preference doesn’t influence the actual result. Don’t ignore the data just because it doesn’t support your hypothesis.

Now that you have officially documented and tested your hypothesis it’s time to run another experiment. Even if the outcomes were positive, think about how you could improve the results.

  • Would you recommend another experiment with changes/optimizations?
  • What could you do differently to improve your results?
  • Example: If the results were positive, could you increase your Facebook ad spend? Could you refine your target audience? Could you expand the campaign duration?
  • Example: If the results were negative, don’t give up on Facebook. Would changing the target audience improve results? Could you run the campaign on different days? Could you improve your ad?

How do you use the scientific method in your marketing campaigns? We would love to hear thoughts in the comments below.