Time-On-Site Well Spent?

By Nate Janitz, Program Supervisor

Question, what is engaging about the screenshot below (screen shot is of actual website)?

Yes, it was a trick question and it highlights the point that just because a person is on a site doesn’t mean that they are being engaged by the site. For years, webmasters have always relied on Time-On-Site to judge how long a visitor is engaging with a particular page. That isn’t a bad idea if your website’s code is optimized to load in milliseconds, but most flash sites are not (nor are most online videos). With rich media greatly improving online experience, truly measuring “effectiveness” of a website becomes even more difficult, especially if the website in examination is built exclusively around content engagement (it isn’t lead gen or ecommerce focused).

No, I’m not saying that time-on-site is no longer useful….it just needs to be tempered with a little logic. For rich media and slow load time sites, engagement-time-on-site is a much better measurement. The idea itself is easy to grasp; unfortunately the formula can be tricky given the tool you are using.
The basic principle behind the idea is that it takes into account the down time in performance when content is not available to the user. Basic Formula is:

T – L = Engagement Time; where T = Time-On-Site(or page) and L=Load Time of Visited Content

Basically you take the time a person has spent on a page and subtract the time it takes to load the page’s content.

Notice though that I am always referring to Content? Take the above screen shot. While technically the page has loaded, the interaction of the content is what we are measuring. In order to measure the interaction, the content has to actually load first. Think of it this way: does an impression count if the person doesn’t actually see the impression? Nope. There is a reason why Offline media pays per impression. Websites should adapt the same mentality.

Great, Engagement-Time-On-Site…So What
While this concept is not limited to just Brand Awareness or Brand Engagement strategies, that tends to be where I use this metric the most often. Let’s say you have a client that has a Brand Awareness site (think League of Clutch by Gatorade). While Query Intent is a great way to measure and optimize the Search campaign, the engagement with the content is the end goal (this works with both Paid and Natural Search). Looking at this engagement-time-on-site can help determine if you need to build out more streamlined landing pages for Paid, tweak text content to target more specific keywords for Natural, or even switch out the baseline technology of the site as a whole.

Try looking at Engagement-Time-On-Site at a keyword level. Developing that critical link between the query and positive content interaction can be difficult. Using the matrix as 1 KPI can help determine tweaks in website that might help improve overall performance.

Tools to help determine load time of Content:
There are several ways to determine load times of a site. Based on the type of content you have, pick the one that work for you. For content imbedded in rich media, I would use one of the “timer” methods. For other sites, I would look at the second list of tools.

Timers that look past load time of the file:

  • FasterFox - A fire fox plug in that helps analyze a load time on different connection speeds. It also has a timer that ticks away as a page loads (think of computer based stop watch).
  • The basic Stop Watch – this isn’t the most technical solution, but you can’t be an old fashion stop watch for clocking anything. The best part is can double as a time keeper for Office Olympics.
Websites that Track How Long a File Takes to Load:
Depending on the site, these three simple websites will tell you how long it takes to load all of the files associated with web page. Just plug the URL of the page in the search field and watch the file load. I usually use Pingdom, but all of them are great for site load times in a pinch.

SomeECards.com says it best, with a card:


CJeffCampbell said...

I can't stress the importance of page load times enough. I've seen way too many high bounce rates as people are too impatient to wait for the fancy-ness to load.

Further, as load time is factored into BOTH paid and natural ranking algos, these fancy designs are limiting ranking success...who cares how beautiful the page is if people aren't getting to it or the media CPCs are too high to even efficiently pay for placement?!?

Copyright © 2008 Resolution Media, Inc. All rights reserved.