Salon owners consistently point to marketing as one of the greatest points of difficulty for their business. When something works, it feels like magic; when it doesn’t, it feels like a mystery. But what if all the time and work you put into creating marketing materials could be rewarded with certainty and insight? What if you could learn what is working and why?

This is the power of experimentation. I’m not talking about simply trying new things. I’m talking about designing structured tests as a scientist would.

Professional marketers utilize an arsenal of tests and analytics to find out what will resonate with their audience, but you don’t need expert knowledge to conduct your own marketing experiments. All you need is a refresher on the scientific method.

The first step is to look at what  you’re currently creating and ask questions about it. For example, maybe you notice that your Facebook posts aren’t receiving a lot of attention: Could this be happening because you don’t include images with your updates? If you are looking at  traditional media, maybe people aren’t redeeming your ValPak coupons. Could this be because the offer isn’t enticing enough? start thinking about what could make your radio ads, emails, social media posts, mailers or website more effective. Every marketing decision you make is made up of many individual components that you could tweak for better results. Colors, fonts, layout, images, offers, calls to action and many other variables contribute to your current success rate, and this step in the process is a time for you to consider which element could benefit from a change.

Once you have a topic in mind, it’s time to form a hypothesis. Your hypothesis is an educated guess about what will happen as a result of your experiment. Usually, these are formatted as an if/then statement. It should describe the action you will take and the change you expect to take place. It is essential that this statement is actually testable.

• If I add an image to me Facebook updates, then more people will like or comment on the post.
• If I increase our referral offer to a full week of tanning, then we will get more referrals.

Next it’s time to design the experiment — this is where things can get a little tricky. If you don’t create the right conditions for your test, your results could be misleading. In the case of marketing, the most common way to test a change is to perform an A/B test. An A/B test allows you to compare two versions of something to determine which version works better. Just like in your middle school science project, you’re going to compare the control (the current version) to the variable (the new version.)

For greater confidence in the results, you should only test one variable at a time,  but you should test the control and that variable at the same time. Email clients like MailChimp make this easy by allowing you to create two versions of an email and send them out to a portion of your mailing list. A portion would get your normal email, while the other portion would get the experimental email. Testing simultaneously eliminates any doubts that the change in results had to do with the time of day or week.

It’s also important to give your experiment  enough time to draw conclusive results. If you base the entire test on one email, social media post or coupon, there won’t be much validity to the data. The more data you can collect, the more confident you can be in your results.

Once you know what you’re going to test, how you’re going to test it, and how long, it’s time to perform the experiment. Follow your procedures closely and make sure they are working. If something goes wrong, check all of your steps and troubleshoot accordingly.

Once your test is complete, it’s time to look at the results. Analyze  the data and draw conclusions. Does the outcome reflect your hypothesis? Why or why not? Sometimes, you’ll find that changing a variable didn’t make a significant difference. Try to find explanations for why the test turned out the way it did, and decide if you will implement the change as a result. Take some notes about what happened, and how the test will affect your approach moving forward.

Once you get started with marketing experiments, you’re likely to get hooked! keep testing to optimize your marketing efforts for all of your business goals. it’s a never-ending process that comes with serious rewards.