The Six Key Steps in Effective Experiments
Now that you have Optimizely set up, it's time to start building your experiments. There are six essential steps that you'll use in every experiment. Let's walk through the steps, one-by-one. This article is an introduction: an Optimizely 101. We'll also link you to more articles that dive deeper into each step when you're ready to explore more.
1. Create a variation: build your experiment
You create variations in the Optimizely Editor. To get to the Editor, just go to the Home page and click New Experiment, then enter the URL of the page you want to test!
If you've already implemented your snippet, you should be able to load the page without problems, but if you run into any trouble, we've got you covered.
In the Optimizely Visual Editor, which lets you make changes to your page in a “What You See Is What You Get” (WYSIWYG) manner, and then test those changes against your original page. Looking for the ins and outs of our visual editor? Read our Visual Editor Guide.
If you're more code-savvy, you can also use our Code Editor to create variations that way.
2. URL Targeting: Where your experiment runs
You’ve got your variation created, so now click Options > URL Targeting to choose where the experiment will run. This feature is called "URL Targeting" because it targets the experiment to display when visitors land on certain URLs (and not others).
Running an experiment on a single page? You'll probably want to use our default simple match.
Changing an element on multiple pages, or sitewide? You'll probably want to use substring match instead.
3. Audiences: Who can see your experiment
Now that we've chosen where the experiment will run, let's decide who can see the experiment by using Audiences. Click the audiences icon, then Create a New Audience to get started. Wondering how to use Audiences? Check out our Audiences article or play-by-plays of each condition you can use.
4. Traffic Allocation: dividing visitors among "A" and "B"
Now that you've decided who can see your experiment, that traffic will be randomly allocated into different variations. You can control how much traffic goes to each variation. You can even decide how much of your total traffic should be in the experiment. Click Options > Traffic Allocation to get started. And if you need more help, check out our Traffic Allocation article.
5. Goals: How you measure success
Now you’re almost ready to start your experiment – but don’t you want to be able to measure results? To do that, you will set up a conversion goal. Ask yourself, what user action represents success? Is it clicking a Call to Action (CTA)? Getting to your checkout page? Submitting a lead form? This is what you should set as a goal.
To set up a goal, click the Goals icon in the top right of the Editor. There are three main types of goals you can set up: pageviews, clicks, and custom events. You can quickly see how to set up a goal in our Goals Overview article.
6. Preview and measure results
Now that you’ve set up all the components of your experiment, it’s time to make sure it works and start collecting results. Previewing is simple: click Preview in a variation tab to open Preview Mode in a new browser window. Check out our Preview Mode article for more information.
Once that’s done, go to edit mode and click Start Experiment. Your experiment will now be live and you'll be on the road to an optimized experience!
You can quickly make changes to your audiences, goals and url targeting more directly from your Home page. Visit our Home page article for more information.
If you have the Optimizely Snippet installed, you’re ready to launch your experiment and collect results! You'll do this on Optimizely's Results Page, and our article on interpreting results gives you the full run-down on how you can view and interpret your results data. Here's a sneak peek of the results page:
Once Optimizely has enough data, it will declare a winner out of all the variations to show you, with real visitor data, which version of your page is best at driving the conversion goal. Aren’t you glad you tested?
That's the end of our Getting Started Guide, but as you create your first experiments and learn more about A/B testing, we encourage you to...
- Check out more articles like this one in our Knowledge Base.
- Then use Optimizely Academy to take lessons that help you move from beginner to expert.
- And be sure to ask questions to the Optiverse Community -- we're all here to support you as you get started, and beyond!