(Part II of a Series on Methods in Web Analysis)
In the last post, I covered some basics about measuring Search. This discussion included some notes on understanding how your measurement solution is capturing (or not capturing) some key search events like loading additional page results, doing advanced searches and repeat searches. In addition, I looked at how to get a few key statistics that will help you evaluate the role of Search on your site. The distinction made there between Search as a Primary Navigation Option (a preferred method of steering visitors to content) and a fallback method (Search only when basic navigation fails) will continue to be central to every analysis of Search. In this post, I’m going to cover Search Sourcing – the analysis of which pages drive more or less volume to Search than is desired.
Traffic to a page is usually generated from a few dominant sources – highly related pages or major navigational elements (like top-navigation). Search (and any other tool available in your global navigation) is different. It will be sourced by lots of different pages on your site. And for tools that are available globally, one of your first analytic directions will always be to understand places on the site that are especially heavy or light sources of traffic. These methods are by no means unique to Search – they are equally applicable to any "global" feature of your site.
You can get the raw counts of Search Sourcing in most tools by using the Previous Pages (or Pages From) reports. This will give you the information you need about aggregate sourcing. What it won’t tell you is how each of these sources compares in terms of their RATE of sourcing. Rate of Sourcing is a simple calculation – Search Sources / Views. Why should you care about the rate? For sites where Search is a fallback mechanism, resorting to Search can be an indication that the content isn’t providing the information a visitor is seeking. Non-Entry Pages with relatively high rates of Search are often content trouble spots for this type of site.
Conversely, where Search is a primary navigation device and the most effective conversion tool, spotting pages where the Search Rate is low can help you identify templates or content where Search is being under-utilized.
To get the Rate of Searching, you just need to match up the sourcing counts (Next Page is Search) to the PAGE VIEW counts for each Source. In other words, if the Products Main Page sources 58 Searches and has 6,100 views, then its Search Rate is just under 1%.
If you are looking at one or two pages, this analysis is easy enough to assemble by hand. If you want to canvas your whole site for Search Rates, then you’ll want to either Export the data and use Excel to match up ‘From Sources’ to ‘Page View Counts’ or you can use your tools Excel integration features (if available).
Using a table that contains page and Search Source Rate, it’s easy to spot places on the site that drive more (or less) to Search than you’d like. You should keep in mind that for many sites the Entry page is particularly likely to generate Searches. If you don’t have many high-usage Entry pages, this may not be much of an issue. But if you have lots of Landing Pages with significant traffic, this can significantly affect your Search sourcing analysis – all of the "Entry" pages may cluster at the top.
Different tools have different methods for addressing this problem. Possible options include Event Sequences, Pathing, and Segmentation. In each case, the basic method is to isolate statistics for the "Entry Page to Search" case and then subtract these numbers from your original Source Counts and Page View Counts to get numbers for when the page is not an Entry.
As with so much in web analytics, there is often some slop here (a page may have been an Entry Page and a non-Entry Page in the same visit and, of course, sements won’t generally capture any numbers from visitors who don’t accept the measurement cookie). For this analysis, however, the numbers are usually clean enough to give you a good working answer.
So how do you use this rate? It’s pretty simple. Establish an average Search Rate for Entry Pages and an average Search Rate for non-Entry Pages. Then compare your actual Page Search Rates to the average.
For "fallback" sites, if a page is Search sourcing significantly more than the average then you need to look at it to see what the problem is. For "primary navigation" sites, the analysis is reversed. Pages that Source to Search less than average may need touching up. Naturally, this analysis isn’t applicable to every page on your site. You don’t want visitors Searching in the middle of an Order Process. As with most analysis and KPI’s, Search Source Rates will benefit from using Functional classifications to make comparative numbers more meaningful. Comparing Search Source Rates with the group of Convincer Pages or Router Pages is always significantly more pointed.
Which brings me to the topic of my next post – evaluating the performance of Internal Search from a Functional perspective. Till next time…
Comments