Internal Search Report

Now, don't confuse internal search analysis with search engine keyword analysis. Internal search relates to the keywords that people use while exploring your site. Search engine keyword use relates to which keywords brought people to your site. ClickTracks does comprehensive analysis of both search engine keywords and internal search, but this article will focus on two important aspects of analysis of internal search.

The Popularity Contest
People search for the strangest things! When you're analyzing your internal search queries for the first time, you'll probably be surprised by the phrases, terminology and common misspellings that visitors use in a search query. While some word combinations are valuable in determining what language and phrases your visitors will feel comfortable with, I recommend that ClickTracks users start with an overview of internal search keyword volume before using segmentation for behavior detail-in other words, get a clear grasp of the big picture view, then drill down.

First, we need to learn the most popular, and the least popular, search terms. In the example below, we're analyzing ClickTracks' demo data site, www.bobsfruitsite.com. Analyzing two months of data reveals some volume for each internal search term.

In looking at the info above, a few things stand out right away.

First: Less than 1% of the visitors used internal search. As a webmaster, I would want to be sure that the search mechanism is appropriately visible and accessible to visitors. It may be worth moving the search mechanism to a different place on the pages and then watch the internal search traffic for increases.

Second: We should question why visitors are searching for balloons and T-shirts on a site that clearly sells fruit. Visitors could be expecting balloons to be available with a fruit basket gift. As the business owner, Bob might increase order size by adding these products to his line of merchandise.

Third: Observe that the most popular search term is "FedEx". From this we might discover that our visitors are having difficulty finding shipping information or, that they prefer FedEx over other shipping companies. In this case, Bob may opt to make shipping information more prominent on the pages that these visitors see most (more on this later).

Are there more insights lurking in that internal search data? Certainly. But for now, we'll move ahead to confirm the last of our observations by labeling and following the behavior of the visitors who searched for FedEx.

Confirming Observations through Segmentation
Now that we know that a significant number of visitors are interested in FedEx, we can label that group and observe their behavior through the site. ClickTracks allows labeling of visitors by a vast number of criteria, but in this case, we will simply label visitors who used the parameter 'srch=fedex'.

If we can see the most visited pages for this segment, we can decide which pages need to include shipping information. ClickTracks shows the most popular pages by segment in the Site Overview Report.

From this, we learn that our FedEx seekers favored our "Recommended" page and our "Organic Fruit" page. If we navigate to the search page, we can confirm that this group came to the search page almost exclusively from these two pages, didn't get what they wanted, and exited the site.

Using segmentation and a path analysis, we know which pages are candidates for FedEx shipping information. It seems clear that Organic and Recommendation pages should offer shipping information to Bob's visitors.

Balloons and T-Shirts?
We can now use internal search labeling to learn if there is some significance to the popularity of "balloons" and t-shirts". Labeling the visitors who searched on balloons and t-shirts will show us if these visitors are serious buyers or just window shoppers.

In this case, 12 of the 29 the visitors who searched for t-shirts ended up making a fruit purchase. Depending on his business model, Bob should at seriously consider adding t-shirts to his product line. Six of the 14 balloon seekers purchased fruit, but at roughly half the value as the t-shirt seekers. For maximum revenue per order, we'd put our priority on the t-shirt aficionados.

What kind of shirts? While some might think it is a stretch for web analytics, we can learn some of the interests of our t-shirt seekers. One glance at ClickTracks Navigation Report can tell us what's most interesting to Bob's t-Shirt seekers.

Looking at the navigation and search engine keywords, we can instantly see that t-shirt seekers prefer organic fruit. More than one third of the t-shirt seekers came to the site seeking using the Search Engine keywords "organic fruit", so it is no surprise that the "Organic Range" page received a whopping 47% of the clicks from the t-shirt segment. We can also see that this segment is more partial to apples than the average visitor.

Also, we can observe whether the t-shirt seekers are new or returning visitors, how much time they spend on the site on average, which geographies they hail from, and a number of other criteria that can help us identify visitor behavior and expectations. The analysis is valuable by itself or as a supplement to market research. The possibilities are truly endless by combining visitor labeling with internal search parameter analysis.

All This by Using Information You Already Collect
Internal search systems typically put the search keyword into a parameter on the URL. Users simply need to tell ClickTracks which parameter indicates an internal search keyword (usually 'srch=', 'q=' or 'search='). ClickTracks derives this information by analyzing your standard web logs (i.e., Apache or IIS), or through a standard hosted JavaScript data collection method -- your choice. The internal search mechanism may also be used to tally other parameter or cookie values which makes it great for determining which product is most popular or which affiliate is truly performing-- all by simply accessing information you're already collecting.

The Internal Search Report is available in the "Site Overview" Report in all editions of ClickTracks version 6 and up.