A good illustration is worth a thousand word. The Botify Log Analyzer report includes many graphs, each with several display options. Here’s how to get the best view to analyze the situation, answer specific questions, or prove your point in a presentation.
The report’s home shows an overview, and you can select the type of indicators you want to see on the right (see open dropdown list below). What we’d like to draw your attention to are the two other choices, circled in red, that are available throughout the report:
- The data perimeter (“all” for the full website in the report home)
- The level of detail shown in the graphs (values minimum %)
The drop-down list on the left allows to select a portion of the website, according to the URL categorization that was defined prior to the analysis.
As the categorization can include several levels of subcategories to meet your analysis requirements, a new drop down list will appear if the selected category has subcategories, as you navigate down the tree structure:
Categories that have subcategories end with “/*”:
Here the URL categorization has 5 levels:
Once you have selected the category you would like to look at, you may need to adjust the level of detail shown in the graph, so that the graph’s message is loud and clear – very useful to get the perfect graph to illustrate your slides for a meeting, for instance.
All selected page categories are not necessarily shown in the graphs by default: only those representing at least 1% of the volume are displayed initially. Which is good to get a first sense of the situation. For instance, here is an example of pages found by the Botify crawler that are also crawled by Google:
If you would like to see them all, then select “all values”:
In this case, the difference between sub-categories is such that we don’t see some of them in the graph (low figures can be found in the full data table at the bottom of the page – and if we want a graph, we’ll need to “zoom in” on these categories, by selecting it).
If we select only categories which represent at least 3% of the volume, the graph becomes:
This functionality is also particularly useful for the “relative” view, which uses doughnut charts, as it becomes difficult to distinguish between many colors shown in the legend, as in the example below:
Of course, when looking at the report online, you can hover your mouse over an area to identify the page category. But what if you want to include the graph in a presentation?
In our example, the graph showing only value above 2% (and the rest as “others”) is much easier to read:
This adjustment applies to all indicators that are displayed by page category, that is to say all but page depth (which shows page depth distribution for all selected page categories).