As website owners, we try to stay on top of things that change from one period to the next. I am always curious to see traffic increases from a specific source to a specific landing page. You know the situation: someone posted a link to one of your pages and it gained popularity. The standard reports in Google Analytics make that kind of discovery very difficult, but it can be easy with a little Next Analytics wizardry.
The actual workbook with the Next Analytics script is linked below, but here’s some insight on how it works. I started with a simple 2-dimensional query: landingPagePath and source, using the visits metric. Then I made the same query for the previous period and joined the results together (see Merging Queries Into a Single Table).
I then change the numbers to a ratio of the total for each column. This results in something like a percent of visits number for each of the two periods. Taking the difference of those columns gives us a delta, so if a combination of landing page and source went from 0.27% of visits to 0.57% of visits, the delta would be +0.3%. Big increases or decreases in this list are of special interest.
Since we have two dimensions, lets pivot the data, showing landing pages versus source. That’s a big table, though, and it contains a lot of what is called ‘sparse data’ where big numbers are interspersed with a lot of empty cells. It would be much better if we could sort the items of most interest to the ‘top’, or upper left corner.
We do that by calculating the range of numbers in each row or column and putting the ones with the widest range in the upper left. I use the range of values because we’re not just interested in the increases – the decreased are equally important (what key referring site has suddenly gone away?).
This report can be embellished with a little Excel conditional formatting and we have quick report that keeps us on top of the trends. Since it uses a Next Analytics floating time period, it can track daily, weekly or monthly differences without having to worry about setting dates – just open the workbook and click Refresh.
The workbook can be downloaded here: two-dimensional-whats-changed.xlsx
Extrapolating on the idea, think what other dimensions or metrics you could use? Monitor goal conversion changes across medium and country? Ad group revenue by product category? Let your imagination go! Simply change the values in the prompt statements for a completely new report.