Onsite Search Analytics - part 1
If you like this post, then please consider subscribing to my full RSS feed
"Basic research is what I am doing when I don't know what I am doing" -- Wernher von Braun
Onsite search is one of the most important features for visitors of your website. In many cases onsite search is your visitor's default behaviour, instead of using the standard website navigation they will immediately browse using the search box and input whatever they are looking for.
Onsite search is even more important when the website contains more data and becomes more complex for your visitors.
Onsite Search Metrics:
Tagging onsite search If you read Recession Proof Web Analytics you have had already got your pages tagged. If not, obviously you HAVE to tag and yes, call who you have to call and tag it since you are probably losing visitors as you read this post, you just don't know it yet. There are some metrics to consider when we want to understand if our onsite search is engaging and if it is really effective as we think it is. The list below is my personal favourite to use as a rule of thumb. Mind you, you would probably need to come up with more metrics relevant to your business later on. Search terms This is rather trivial. The data set should include all the search terms that were submitted via your website search box. When sorting this data set in descending order you will get the most popular search terms on your website. Search terms per page This is where it starts to get interesting. For some reason, online businesses often assume that visitors will be only searching from the Homepage, which is so far from the truth. In reality, visitors are submitting searches from different pages. If your website was tagged in a logical directory structure you will get some invaluable insight on trends in onsite search with regards to directories and pages. Depth of search Say you sell commodity products on your website. A visitor is submitting a search and gets a result set back from the server. The result set is relatively large and might be divided into more than one page. If you followed the previous recommendation then your search pages have already been tagged, but we could get some more insight if we tag the number of the results the visitor got and the way they engage with the result pages. "Why do I need that information" you ask? Many of your visitors do in fact find what they were looking for but it may be on the second or third page of the result set. Wouldn't it be great to add these pages to your Best Bets mechanism so the next time a visitor searches for the same thing they will actually find it on the first result page, instead of leaving the site because they can't be bothered to navigate through the rest of the search result pages. Zero result search terms This data set should include all the search terms that visitors submitted via your website search box which resulted in zero results. Zero yield search terms This data set should include all the search terms that visitors submitted via your website search box which resulted in data that was returned from the server but the visitor never clicked on a single item on the result pages. Search terms conversion This data set should include all the search terms that resulted in a conversion. So far so good. Now we have we have 6 reports that we can run in order to get more insight with regards to our website and on site search. Can we derive REAL KPI's from that? Sure we can, just continue reading the second part of Onsite Search Analytics.
Related Posts: (by tags)
