GA Site Overlay (and a letter to you, Avinash)


Dear Avinash:

I know you say that you aren’t the guru of Web Analytics. Nevertheless, every time you speak or write, people change how they do their work based on your thoughts. Some of those people are my customers!! So you have to start approving your posts with me first Website Worth.

Well, I guess that won’t happen. But here is my concern with your excellent blogpost today: Unlike you, I never look at the site overlay, and I think it is the least helpful report in GA.

Here is why: the Overlay is just a visual representation of the content > Top Content report. So if you have a link to your shopping cart on your home page, and you have a link from all your other pages, they all get the same values in the Google Analytics overlay. There is no differentiation based on which page the visitor was actually on when he made the click. You can add a query parameter to differentiate links that go to the same page, like so: /shoppingcart/index.html?link=home. However, you (arguably) create an SEO problem for yourself by creating duplicate content with that technique.

Furthermore – as soon as you rewrite your URLs in any way, your overlay breaks. Here are some examples:

  • You want to see the entire url, not just the stem. This is very helpful when you have subdomains. But when it gets written like, the overlay breaks
  • You want to rewrite your urls so that you can see them ALL by page title. Do the rewrite, the overlay breaks.
  • You rewrite your pages so that they pass a value to urchinTracker in the old Google Analytics, or to trackPageview in the new Google Analytics. Now you have a nice friendly name, but oops! your site overlay is broken again.

We can get around some of those problems by having a profile with no rewriting. But when you do the rewrite in the code (like passing values to the urchinTracker / trackPageview example), that isn’t an option. And still you have the problem of not differentiating links among pages. So I never find any use for the overlay (and am curious to learn about who does….)

Our founder, Robbin Steif, started LunaMetrics in 2004. She is a graduate of Harvard College and the Harvard Business School, and has served on the Board of Directors for the Digital Analytics Association. Robbin is a winner of a BusinessWomen First award, as well as a Diamond Award for business leadership. In 2017, Robbin sold her company to HS2 Solutions and has since retired from LunaMetrics.

  • Dustin

    100% agree with you on the GA Site Overlay. When I first used it I thought it to be beneficial, but once I realized no mater what page you are on the links always have the same value, it becomes non-beneficial. I am not really sure why it is even on there for a report it can definitely give people the wrong idea if they don’t realize what it is actually reporting.

  • I agree.
    I basically find it useless (for more or less the same reasons as specified). If I’m in need for visual click pattern analysis, I always return to So unless GA does something like their reporting they might as well leave the site overlay out…


  • Yes! Great post. I couldn’t agree more, and have always hated the site overlay! Until GA differentiates between links to the same url on a page the site overlay will be eye-candy at best.

  • Dear Blog Post Author:

    It is my sincerest wish that we can separate the methodology from the tool.

    My Web Analytics Demystified post was targeted towards new users of web analytics. I am utterly convinced that, whatever the tool, Site Overlay is the best methodology to understand what is happening on pages on our websites.

    Yes between the page data report and the navigation report and the referrals report and next click report and …… I can see exactly where people come to a page from and what they do. But the Site Overlay is the most easy to use, and damn visual to boot (!), way to get that data from one page.

    On to the tool…. Some tools are good at doing Site Overlay and other tools are not very good. If the tool you have, or I have :), does not work then try something else. Trust me people who make tools will notice! I shared my love for ClickTracks in the post. I also shared the confetti screenshot of CrazyEgg in my post (which Mailand points to in his comment).

    In my book Users come first and I am sincerely appreciative of all the feedback!


  • Dustin and Mailand and Nicholas – thank you all so much for writing. (Even if you had disagreed, it would have been great to hear from you.)

    Avinash — I know that you can separate the methodology from the tool. But I cannot. GA is the most important WA tool we use at LunaMetrics. We do use Crazy Egg, and it is wonderful for site overlay abilities, and it really helps me understand how a page is used — but in a way that the GA site overlay does not.

    Maybe we can just say, “Overlays are awesome, and the GA overlay sucks.”

  • steve

    Interesting discussion. I’ll sit more on the fence on this one. Don’t want to be the next target of Robbin’s Unleashed Wrath. 😉

    1. I find the Site Overlay feature a god send for being able to clearly demonstrate to the system owners which parts of the site are redundant or simply not used. They’ve *known* this for years – but graphically showing it (and be honest – attached to the Google Magic Brand) almost made them take action at long last. Key word is “almost”. 🙂

    2. OTOH; I had no idea of the problems you raise Robbin! That puts a severe damper an any enthusiasm I have had.

    3.With Work’s site: we have lots of pages that are fundamentally identical – different contents, but same structure. The site overlay has a very 1 to 1 correlation to pages, so it’s very hard to answer for real life example: “Just how useful, site wide, is the Right Hand Side Nav Bar”. I’ve stuck with my funky perl scripts and log magic for that one.
    Knowing that the RHS Nav is no good on the “Blood Transfusion” page (for example) is of very questionable value.
    It just doesn’t scale well when you’re talking ~1200+ key pages.

    Perhaps I should look at crazy egg again…

    – Steve

  • I don’t agree on this next sentence:

    “Here is why: the Overlay is just a visual representation of the content > Top Content report.”

    In my opinion the Overlay is a visual representation of the Content > Navigation Summary. If I check the Overlay on Page A I see 10 people that clicked on the link to C. But if I check it on page B I see only 5 people that click the link to Page C.

    Avinash and Blog Post Author: what do you think?

  • Hi Robbin,

    I too didn’t find Site Overlay very useful when all the major vendors started to imitate ClickTracks on this. In fact, it can be very useful when one analyzes how a page, seen as UI, redirects traffic. The problem you mention is common. However, in more sophisticated applications than GA (WebTrends for instance, which I know well), one can add code to the links, so that one can know which of the many links to the same page got the clicks. Or maybe it’s possible to do it with GA; I dn’t know.

  • Hi Jacques! in fact, we can add that same code (like WebTrends) to the GA link, it only creates SEO problems.

    Andre, you do such a nice job on the GA Groups, you and ShoreTel deserve awards. And to your point here – I found a profile where the overlay wasn’t broken, and you are *right*, the links are no longer identical. So now I need to go do some testing, and maybe I will have to rewrite a portion of this. I haven’t used the overlays in so long, maybe they have changed ten times without my looking at it…. I have to do some testing.

  • But I am definitely going to start signing my posts…..

  • steve

    “But I am definitely going to start signing my posts…..”

    FWIW they are in the RSS feed, just don’t show up on the page itself. I suspect there’s some sort of template “put here” that could be used.

    – Steve

  • Aaron

    Thank you for addressing this issue. I am very interested in hearing Avinash’s response. I am presently reading his book and have also wondered about his recommendations about site overlay.

    I have no doubts about the POTENTIAL value of site overlay–in terms of showing us where user’s are clicking–but I see major technical challenges in implementing an overlay that is accurate.

    Some of the diagrams in Avinash’s book show “hot spotting,” not just site overlay. I think a true hot spot visual would be helpful but I wonder how it is implemented.

  • I definitely agree. Site overlay sucks big time.

  • Pingback: eTc :: El blog de Marketing en Español » Blog Archive » Para meterse en la mente del usuario: Site Overlay()

  • well Avinash is very Sharp About this, but where is his post regarding this post,
    i am waiting for his reply,

  • DrsMgmt

    I think GA SO is marginally helpful, however, I have a unique situation.
    Our website navigation is terrible and very convoluted due to a content mngt system that too many people have access to (I’m here to change that…have been w/the company a month).
    We have so many layers of navigation I’m a bit dizzy. MANY of our links show up as 0% usage in GA SO.
    I think this is helpful at least for the first go-around at redesigning our website (which is currently over 2800 pages!!!).
    I appreciate the discussion as well as the ideas for other tools/options.

  • Nice post & good discussion, but the main issue of overly is missing.

    Roocky 🙂

    • Robbin

      The site overlay is crummy. There isn’t much more to say – Robbin

  • Google has removed the “Site Overlay” tool from Google Analytics and replaced it with a new one called “In-Page Analytics.” It’s fairly similar to Site Overlay, but appears on first glance to offer better data visualization and an improved interface.

Contact Us.

Follow Us



We'll get back to you
in ONE business day.
Our Locations
THE FOUNDRY [map] LunaMetrics

24 S. 18th Street
Suite 100

Pittsburgh, PA 15203


4115 N. Ravenswood
Suite 101
Chicago, IL 60613


2100 Manchester Rd.
Building C, Suite 1750
Wheaton, IL 60187