Analytics as we know is fantastic at reporting on many facets of your online ventures. However, one little nugget of wisdom is “Analytics can only report on what it can record” (courtesy of Avinash Kaushik). If you haven’t set up Analytics to record properly, then the data you receive is going to be patchy at best.
There are many areas that you can tune Analytics, and we are going to look at one in particular: internal site search.
This is valuable search information as it shows what the customer is looking for after they come to your site, on potentially an a search term that may be largely ignored in your account. The data that internal search provides can be huge. If you gather the data, and find that people are searching a term that you do not have within your PPC account, then there is actionable insight right there. Analytics is all about outcomes after all. Quantifying the data you are receiving and evaluating your next steps to achieve outcomes. This is the way forward, not looking at individual KPI’s exclusively on their own.
Recording internal site searches
So how do we get Analytics to record these internal searches? As easy as a tick box.
Navigate towards your settings which is the cog wheel at the right hand side (these screen shots are for the new Analytics) then edit the profile.
Once you are there, tick “Do track site search”. Within the query parameters, enter the search terms that are a constant when you do a internal search. For example, if you do a search and the parameter is, search=blue widgets, or q=blue widgets, then you can enter search or q as the parameters.
This enables Analytics to track these separately in your reports within the Content -> Site Search segment.
Once this has started gathering data you will start seeing search terms appear. There are some interesting statistics that this type of segmentation can offer you. For example, we can see how “deep” the user went after searching for a particular term.
In this example, the user searched for “fender” and stayed on the site a further 5 minutes
Interestingly, in this example, there was 11% search refinement. This means the user refined their search query further as the internal search engine potentially did not return the most accurate results. You can use this sort of data to present the case for improving your internal site search.
As ever, we would be interested in hearing how you find using this and what data insights you get!