How to determine influential pages with Google Analytics

Let’s say you are responsible for data-driven insights in your company. One day your boss comes to you wanting to know what is the most popular content.  ‘Doh’ you think to yourself ‘that’s pretty straightforward’. And little fun.

As a mighty web analyst you want to go further and understand whether the most viewed content is actually most successful in driving leads (or other types of conversions).

In other words, you want to:
– understand what pages are driving most conversions
– see how you can act on that data

To evaluate the content effectiveness in terms of conversions we will need to calculate the view-to-convert ratio for each page.

Steps to calculate the view-to-convert ratio with Google Analytics

1. Run the All Pages Report for the appropriate time frame and export to Excel the URLs and the Unique Pageviews for each one.

This will provide you with number of visits during which the specified page was viewed at least once.

Tip: The reason we are not using the Visit metric is because in Google Analytics Visits means Entrances.  GA’s way to show visits is through unique pageviews. But you already know that 🙂

2. Create and apply the segment where the session contains the conversion page. Refresh your pages report. Export to excel the URLs and the Unique Pageviews.

This report will give you the number of visits where the user both viewed the page and converted.

Tip: Traditionally Google Analytics didn’t consider whether a specific page preceded or followed the conversion, as long as that page was visited during the session. Recently Google Analytics re-imagined segmentation (July 2013) and now you can slice and dice  visits according to sequential conditions that suit your scenarios.

3. Combine both reports into one in excel so that each URL has both metrics.

Tip: You will need to use vlookup function for this one (this guy here covers it awesomely http://www.youtube.com/watch?v=-WAEzokHSJM)

4. Add an extra column with Conversion Rate for each page.

(Conversion rate is Conversions/Unique Pageviews)

5. Clean the data and eliminate the noise.

Delete the URLs that are part of the funnel, have low instances, etc.

So what do I see here?

Pages that have most visits do generate most leads, but the conversion rate of most of them is mediocre.

Actionable takeaway: By improving their performance there is definitely potential to increase the revenue.

(note: I changed the labeling from unique pageviews to visits)

I also see that content with good conversion rate has little traffic.

Actionable takeaway: Understanding why so few visits reach that content and making it more accessible can generate more leads.

In my case I wanted to find what is the most influential content but this principle works with determining most influential products!

Determine the view-to-buy ratio of your products. You might find that some of your high revenue products are not as profitable as more hidden ones. You might also discover common trends between most or least popular products.

Having this data is only half the work in terms of analysis. In one of the next posts I will cover how you can better visualize this Data. It will help digging deeper and on a larger scale (great when you have a lot of traffic and pages) and, more importantly, some ideas on how you can act on that data for better business results.

Considerations

The approach I’m describing above can be both session-based or user-based. Use what works best for your scenario.

Now I want to hear from you. Tell me what you think by posting in the comments below.