What is a Page anyway?
In a traditional sense, a “website” is equivalent to a domain (like foo.com), and a “page” within that site is equivalent to a single specific URL (like foo.com/about). For many content-centric landing pages, or basic websites, this definition makes a lot of sense!
But today, most modern websites and web applications do a lot more than just serve static content pages at static locations. Their URLS are more sophisticated, often dynamic, and generally contain data in them.
For example, this URL:
is really an instance of a “Product Detail Page”
When analyzing interactions with your Product Detail Page, you usually want to look across all Product Detail Page URLs. You also don’t want to mix in interactions from different pages like the “Search Results Page” or your homepage.
Pages in FullStory
To scope your page-centered analysis to the right level of granularity, FullStory uses machine learning to identify general URL patterns across similar page structures and group them into Pages. This doesn’t require any instrumentation on your part. If you see brackets [ ] or asterisks * in your Pages, that means that multiple URLs are grouped together into a single Page.
You will see a list of all your Pages in Settings > Data Management > Pages. Those created by the FullStory machine learning algorithm, called Learned Pages, will have a purple ML icon.
You also have the option to override these Learned Pages by editing them or creating new ones.
Note: Users with Admin or Architect roles can change their names and/or URL rules. To learn more about editing your Pages, see Page Settings.
These Pages are used for search and page analytics across FullStory, including Visited Page searches, Click Maps, Scroll Maps, Segment cards, and Page Flow Dashboards cards.
Because FullStory is constantly learning about your site’s URL and page structure patterns as your site is updated and traffic patterns change, Learned Pages will evolve over time which may cause data fluctuations for a given Page. The data you see is always using our best fit clustering at the present time, based on historical trends. FullStory checks for new or updated Pages nightly, though it is rare your Pages will change that often.
Example of how Learned Pages work
Imagine you are shopping for blenders on ecommerce-co.com. You could see two different URLs for blenders with different words on the page, but this page is the same Product Details Page.
FullStory uses machine learning to know which page metrics should be combined across certain URLs by matching their URL pattern and page UI structure, even if the content on each page is different.
The fact that multiple URLs have the same UI structure allows FullStory to interpret what parts of the URL are variable. Using that intuition, we can turn these URLs...
...into this Learned Page:
Frequently Asked Questions
What does it mean when I see an Unknown Page?
When you see the Page is Unknown, that is grouping together page views for URLs that are not already defined as a Page. By default, FullStory will create Learned Pages for your most commonly visited pages. We don’t automatically create Pages for all of your URLs so that your experience isn’t cluttered with Pages you don’t care about.
In order to find out what URLs are getting grouped into Unknown, you can create a Dimensionality card in Metrics or Dashboards for events where the Visited Page = Unknown, and Grouped by URL.
You can add more Pages to capture that additional traffic in Settings > Data Management > Pages by clicking on “Name Page.”
Why do I see adjusted time frames in Page Insights and Page Flow cards?
When new Pages are created either by the Learned Pages algorithm or by someone on your team, their data is captured on a go-forward basis only. Therefore, if you look at a Click Map, Scroll Map or Page Flow card for that Page and your original time period starts before the Page was created, FullStory will let you know that the data actually only goes back to the day the Page was created.
Can a URL be mapped to multiple Pages?
No, URLs will only get counted towards a single Page. This allows you to compare and sort your list of pages without worrying that some pageviews are double counted. In the case there are two Pages that could match to a URL, the more specific one will win.
For example, if you have these two Pages:
- Product Details Page `https://ecommerce-co.com/p/*`
- Product Details Page - Special Blender `https://ecommerce-co.com/p/special-blender-GHJAVMY/`
Any visits to `https://ecommerce-co.com/p/special-blender-GHJAVMY/` will only be counted towards the Page Product Details Page - Special Blender.