How do you track purchases based on click at start of funnel? (free)
Pardon the newbie question, but I'm currently testing the FREE plan and am wondering how to achieve the following:
I'm testing different CTA wording for my Buy button. Besides counting button clicks on the starting page, I also want to know how many users complete the buy process (which is several pages long on another domain). How can I measure the latter per variation? The final purchase confirmation page sits on another domain, but I can put my optimizely code there if necessary.
It can be as simple as seeing how many ppl reach that final page per variation, but I can't seem to find how to set that up.
@kevin1 - Welcome to the world of Optimizely.
For your landing page, you'll want to set up a "Click Goal" which it sounds like you have already done.
For conversion events, you will want to set up a "Page View Goal" that records when a user hits the URL of the thank-you page (and, yes, you will need to have the same Optimizely snippet loaded on that page)
Analytics and Testing Guru
@JasonDahlin - Thank you for the reply, but how do you tie conversions on the final page to clicks on the landing page? For example, my landing page is tracking clicks on 4 variations of the buy button and I want to see which one leads to more people reaching the final page (conversion event). How do I attribute conversions on the final page to each variation of the buy button?
Optimizely will automatically attribute those different buttons to the result. As long as each button is a different variation then it will attribute properly. If I understand correctly, you are wondering about the targeting and goal settings.
Optimizely can track page visits on URLs that your experiment is not running. So you will run the audience settings to the page with your buttons (I assume a simple match of that URL), then track visits to the success page. Optimizely will attribute that visit back to the variation, there is no code you need to follow because Optimizely recognizes the cookie of the variation that it sets when a visitor first enters.
On the free account this is the option that is availible to you and will require more traffic because you won't be sure that the person actually clicked the button. What you may be looking for is Conditional activation which would only trigger the start of the test based on a click of the CTA. I believe this is a paid feature though.
Mike St Laurent, Optimization Strategist - WiderFunnel
Thank you for the reply! But what do you mean by "you won't be sure that the person actually clicked on the button"?
And just to clarify about my experiment: The starting page URL and the final purchase page URL are the same in all variations. The only thing I am changing is the text on a single button on the starting page. The button still clicks through to the same purchase funnel in all variations. I already setup click goals to track clicks on the button variations. But I want to know which of those variations will lead to most page views on the final purchase confirmation page down the funnel.
The Results in the Optimizely dashboard show you the breakdown of how often the goal was achieved (seeing the thank you page) by all members of the audience (people in the experiment). If there is another way to reach the thankyou page, then those entry point will also be counted in the results. If there is one and only one way into this funnel and that is the page that you are modifying, only then will everyone be a "CTA clicker".
Consider - you are running a CTA experiment on an SEO-specific landing page. A user sees Variation B but decides to click a link to the homepage, do some research on your site, then places an order. This user never clicked on your CTA but they are in the audience (they saw the experiment) and they placed an order - their conversion counts.
This is not necessarily a bad thing, it's just different. A user can see an experiment and convert, but not actually click on the modified CTA. But, users who see the CTA are influenced by it. It may be worthwhile to create a custom segment (dimension) for users who click on the page's CTA. You could then look at your experiment by CTA-clicker conversion and non-CTA-clicker conversion.
Analytics and Testing Guru