Project: Goals and Funnels

For this project, I created a scavenger hunt on this user journal with clues for users to follow in order to visit and scroll down 5 pages. The pages are in a variety of formats, some blog posts, others pages reachable from top navigation. Content is varied as well, including images, video, and a link to an online game. All clues are within one click away from the answer.

GA-SH

Set Up

Google Analytics

I set up a destination goal using Google Analytics to track user progress onto the landing page (the blog post with instructions shown above) of the scavenger hunt.

I also created a funnel with the order of pages users would take if following the clues sequentially to the finish. I chose not to make the funnel required, because users may take different routes, or become interested in other content along the way. Reaching all of the pages, not necessarily in order, is more relevant to completion.

WP GA Events

To compensate, in part for not making the funnel required, and in part due to my own fascination, I used WP GA Events to add a scrolling event to each page. The id for the scroll is located in the clue  in order to observe how many users scrolled to the clues on the page. In addition, I added a click event to a comment link on the final page to see how many users would return and comment from the link.

Implementation

Once I created all of the content, coded scroll/click events, and set up the goal/funnel, I sent invitations to some of my personal contacts to drive traffic to the landing page.

Initial Findings

Scroll Events

Below are the top scroll events ranked in order.The events are ordered by clue number (clue 1 is on the landing page).

GA-SH

 

From the order, I can already see that pages with clues 4 and 3 show more engagement levels than others, which may be due to users returning to the page more than once. What is interesting is that there are 19 events tracked on the landing page, but only 6 on the page with the final clue.

Goal Completions

At this point, there have been 9 destination goal completions tracked to the landing page.  Five of the total originate from Facebook where I invited personal contact with a direct link to the landing page.

GA-SH1

Of the 9 total conversions, all but one entered the site via the landing page, which is what I anticipated. The 8 direct entrances are consistent with my main traffic source via the link I sent to contacts.

GA=SH2

Of these 8 conversions, only 2 continued on to the next page in the scavenger hunt to view clue 2.

GA-SH-flow

At this point, they are joined by 3 users arriving from the home page of the blog who are traveling along to the contact page and clue 2 as well.

GA-SH-flow2

There are 0 drop off from this group of 5 until they begin to reach the end of the course of scavenger hunt clues. At that point 2 drop off and 3 continue to travel the site. As you can see from the images 1 is returning to the ‘play the game’ page again.

GA-SH-flow3

Others diverged from the path slightly and are still completing the hunt at this point. In fact, it appears that at least 2 users are going through the pages a second time.

GA-SH-flow4

All users of these users drop off by the 11th interaction on the site (final pages were Contact Us and Play the Game).

GA-SH-flow5

This users flow through the site supports my earlier assumption that some pages are receiving higher scroll event results due to return visits to popular pages.

Conclusion and Next Steps

I’m pleased with the initial data collection results. They are rich and varied, which give me the opportunity to examine how different measurement methods reflect user behaviors.

One part of the hunt that I didn’t track at this stage was leaving a comment upon completion of the hunt. Only one user completed this task, which may be due to the fact that it requires returning to the landing page to enter the comment (or that users don’t want to leave a comment).

Moving on to the next phase of our project, I plan to take a look at the content and organization to determine what improvements could be made to improve overall results, but also improving ease in commenting at completion.

Leave a Reply

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