Page Names in Analytics and why they are important

Page Names in Analytics and why they are important
For those of you who are unaware, I have recently changed jobs and in those situations you always end up thinking about new things in different areas.  Recently one of the things that I have been thinking about is how you set up naming conventions in your Web Analytics tools (for those that need naming conventions).  It’s something that I’ve thought about before (obviously), but only consequentially.  So here I thought I’d give some generic advice on what you should be looking at when you come up with some naming conventions.
Page Names Must be Unique
Sometimes I’m not convinced that people get this one.  Of course if you are using Google Analytics all your page names are already unique because they are the url of the page by default, so this makes life a little bit easier.

Of course there are downsides to this approach that aren’t immediately obvious by looking at the urls of this blog:
  • Firstly there is a distinct possibility that there may be additional query strings that are associated with a page that you aren’t aware (for example session IDs).  It’s unlikely with a blog, however just as simple search for the most visited url on this blog brings back two other entries in my Google Analytics for versions of the page with strange query strings that Google hasn’t ignored.
  • Secondly it is possible that the same page may appear in various different forms of a url (eg and  This is obviously bad from an SEO point of view (you only really want to have one url per piece of content), however it is possible to exclude one from the search indexes (or use the canonical url parameter), whilst still having the same url.
  • Thirdly it is possible that your CMS can populate pages without changing the url of the page.  This is particularly true for situations where the user submits a form and the next page says that you’ve completed the form successfully.  This is somewhat annoying because these are normally the pages that you are most interested in!

There is a way around this, of course, because you can get your developers to use the _trackPageview() parameter so that you can specify what you want to collect as your Page Name.  If in doubt you could use your canonical url parameter.
In other tools, the Page name variable is something that you are advised to set in the code (even if you can leave it blank in tools such as SiteCatalyst to pick up the URL).  So this gives added incentive at the start of your implementation process to ensure that you are capturing it in a unique way.
When in a situation where you are having to specify the name of the page, then you can also come up with a couple of problems:
  • You may end up with two identical pages having different names depending on how the page is built (using our crude sports socks example above you can have them both way rounds and end up with a Sports:Socks and Socks:Sports page name with the same results).
  • Capitalisation tends to be very important and you should do everything you can to put all the Page Names into one case.
  • If you are using Page Title tag, many pages may have the same title.  This will mean that your pages names are no longer unique.  In fact this is one of the areas that it is sometimes the easiest to give pages unique names as you can assign CMS IDs to page names.
Of course there are ways around all of these problems, but you might need to make sure that you look at some real examples before you aimlessly push these things live.  Testing in environments which don’t mimic your live environment in terms of content may mean that you don’t pick up one or more of these.
What information can you get out of these reports?
Pages reports tend to be the reports that allow you to do the most pathing analysis.  This is an important bit of knowledge when it comes to looking at your site scent.  Having the ability to know what people did having arrived at the site is one of the most important things you’ll be able to do to increase conversion (or page views, or whatever it is that you want your users to do).
Don’t forget that you’ll probably also want to use unique pages in your funnel reports.  This is pathing at its beautiful best.  You get to take a series of users’ paths and work out how far they went through your process before giving up (or going elsewhere).  This can allow you to focus your optimisation skills on the areas where you get the biggest benefit.
But it is not just pathing reports that are important, it is those bookends of your pathing reports that are really important: Your Entries, Exits and Single Access (or Bounces).  Having the ability to know where a user entered the site and where it exited is really important in working out what you can do to improve the site.  Everyone has to Enter the Website and Exit it, but they don’t have to do all the other stuff in between!
Without having unique page names (or too unique if you suffer from having different page names for the same page) you may end up misinterpreting your reports and spending too long trying to optimise the wrong page or not enough time trying to optimise those pages that would benefit most.
The other thing that these reports are almost always linked to are any link tracking technologies.  What your system will do is track links against the page that they are on.  If you suffer from non-unique pages you may end up confusing these reports and wasting your time trying to work out how users clicked on non-existent links or why data looks different today than it did yesterday (when you were looking at a different page name in your Analytics, but the same content).
What Should be in a Page Name?
The million dollar question about page names is always: “What should I include in the page name?” and there is no easy answer to this question.
I think it is always useful to remember what the reports look like and what metrics are available.  Remember that you will have visits (or unique page views in Google Analytics) which shouldn’t be added together.  Just because you can’t add them together doesn’t mean you can’t count them elsewhere in a different custom variable in a different grouping!  A large proportion of the time you will be looking at the page individually (hopefully you’ve made them unique!).
Sometimes however, you may want to group some pages together to look at other metrics that you can be added together.  With entries, exits and single access, you can add your values together to get a sensible answer (you can only do one of these per visit so there is no need for deduplication).  So you’ll need to think about what you can add where you would need the grouping.  This might mean that you want to look at particular sections, types of pages, etc.
So hopefully that gave you some food for thought. It’s important to make sure you think about these things early on in the process to ensure that you don’t have to revisit your development processes to get the tagging right – you’d prefer to be able to spend that development resource making the website! 

6 Comments on “Page Names in Analytics and why they are important

  1. Good post, but when you say “There is a way around this, of course, because you can get your developers to use the _trackPageview() parameter so that you can specify what you want to collect as your Page Name.” this is not exactly right… _trackPageview can take only one parameter, a URL to be used instead of the default. Even with this option, it will still use the same page title (the TITLE value).

    To work around this and really specify something else for the page title, you can use enhanced _trackPageview. And as a bonus, you can even track outbound links & downloads!


    P.S. Disclaimer: I created gaAddons, which is being used by a growing number of agencies.

  2. Hi Stephane,

    Thanks for the comment!

    You are right – you can only change one the one parameter. This will probably be enough for those who have the odd accidental issue with things like a form that doesn’t change url.

    For those looking for more than that should check out!


  3. I do web analytics for a large newspaper conglomerate. I have a JS call that initiates a JS page code script. The problem is I have trouble making sure that the content providers at the papers tag the pages. I need a way to index a site, and see if the JS is included in the source code. Normally I go page by page and view source, then ctrl-f and search for the name of the script. Does anyone know of a program that will do this for me?

  4. Hi Marcella,

    There are a couple of tools that you can use to check your web analytics tags.

    Stephane created a tool called WASP which identifies which analytics tags are on each page, along with the parameters that are captured. This tool is free to use for 20 minutes, but does allow you to crawl the site if you upgrade to the premium version.

    If you are using SiteCatalyst, my company (Adversitement) has a tool called WATS which is free to use, but a bit more limited than WASP as it only does SiteCatalyst. It’s also not automated, so you’ll still have to go through the site page by page.

    Hope that helps!


  5. Good Post, thanks!

    I have a quick question. There are times when after clicking on a web page the URL does not change but that information is very important for analytics. DDo you know how can it be captured if the URL remains same (on JSP).


  6. Hi Kumar,

    It depends on the tool you are using, but you can make your page name be anything you want – the url for SiteCatalyst and Google Analytics is just the default one.

    In Google analytics you can use the custom code on the page with the following (which should already exist on the page but with nothing in the brackets):


    In SiteCatalyst there is a s.PageName variable that you can set on the page to be anything you want.

    In all the other tools there is an equivalent. The trick is that you need to slightly customise your implementation to make it get the data you want out of your site.



Leave a Reply

Your email address will not be published. Required fields are marked *