What Google Analytics can't tell you . . (part 1)

RandyNorton99 · 10-16-2009, 12:12 PM

Time on Page and Engagement Time™

We love Google Analytics, we always have. Just like ClickTale it’s free, easy to set up, comes with a lot of helpful tools, and is a great way to collect analytical information about your site. However, there are several things Google Analytics just can’t tell you, and in this post we’re going to talk about two of them:

1.Google Analytics cannot tell you anything about bounced visitors! These are visitors that come to your site, only look at the page they land on, and decide to leave. These are the potential customers you didn’t get, they are the lost sales, lost leads and lost profits.

2.Google Analytics gives you no information about how long your visitors actually interact with your online content. All it can see is the amount of time a page was left open, which doesn’t tell you anything about how long your visitors were actually looking at your content.
And now we’re going to tell you why…

Like most traditional web analytics services, Google Analytics records a “Time on Page”, denoting the time a visitor spends looking at each page in your website. It does this in three stages:

1.It records the time your visitor opens the first page.
2.It records the time your visitor opens the next page.
3.It subtracts these two times and calls the result “Time on Page”.

This method has been used by Google since it started it’s analytics service back in 2006, and while it is a simple way to gauge user interaction, Google Analytics’ method for calculating “Time on Page” and “Bounced Visitors” is woefully inaccurate!

OfferUK · 10-22-2009, 09:53 AM

Analytics can tell you quite a bit about bounced visitors - it's often extremely useful to monitor people who are leaving your site immediately.

JennyRipley · 10-23-2009, 04:04 AM

I don't agree your first point.I think it can tell you a bit about bounced visitors.

SEO Dave · 10-24-2009, 08:18 AM

Very true. It's inaccurate in calculating bounce rate, as it doesn't take into account the time on a page unless the user clicks another page = inaccurate. I also use clicktale to measure those "google bounces" and it always varies a lot. Some searchers can find the answer (or not) on the landing page - be it a telephone number, answer to a question (think wikipedia - how many poeple click on a second page if it's all there?) so it's not really a bounce.

udayns · 10-27-2009, 02:19 AM

Google analytic can not show the exact time of visitors.

moonheart · 10-27-2009, 02:24 AM

I am using google analytics on my blog. Their features are really very nice. Now they have install one new feature where you can see your backlink with anchor text.

luu · 10-27-2009, 07:19 AM

google analytics cannot tell you the exact time that visitors get into your website

**~kev~** · 10-27-2009, 02:09 PM

Quote:

Originally Posted by RandyNorton99

This method has been used by Google since it started it’s analytics service back in 2006, and while it is a simple way to gauge user interaction, Google Analytics’ method for calculating “Time on Page” and “Bounced Visitors” is woefully inaccurate!

And your point is?

I fail to see a problem with the way Google Analytics reports traffic. In fact, I think its pretty good.

When your site gets 10 hit a day, sure you might have time to look at everything. But when your getting thousands of hits daily, and millions of page views monthly - its a waste of time to dwell over little fine, small numbers. That i why Google Analytics puts everything together in averages.

mradamsmith · 10-27-2009, 05:00 PM

You know what it cant tell me, visitors on a certain day in a certain hour, which pisses me the hell off. If you select multiple days then yes, but for example today from 10 am to 11 am, no way to get that info. Which sucks for high traffic sites.

mr.joebert · 10-27-2009, 07:38 PM

Quote:

Originally Posted by mradamsmith

You know what it cant tell me, visitors on a certain day in a certain hour, which pisses me the hell off. If you select multiple days then yes, but for example today from 10 am to 11 am, no way to get that info. Which sucks for high traffic sites.

I can. I'm looking at the hourly statistics for yesterday right now.

Select the day you want from the drop-down calendar on the right.
Select "Visitors ->Visitor Trending -> Visits" from the sidebar.
Select the thing that looks like a clock next to the words "Graph By" on the right of the resulting page.

Not all that useful for todays statistics, but from yesterday and back it works just as I would expect.

jason830 · 10-27-2009, 08:33 PM

I think google analytics is good, i can find some problem with my site from this analytics.

**juliensimon** · 10-28-2009, 12:43 AM

If you have a link in the footer and a link in the top navigation menu, Google Analytics cannot tell you which one people clicked on.

You need a heatmap for that.

mr.joebert · 10-28-2009, 07:53 AM

You could use a heatmap, or you could just use the Event Tracking API of Google Analytics to traverse through the parentNodes of the link and attach the first ID attribute it finds to the URL when tracking the event.

RandyNorton99 · 10-28-2009, 11:12 AM

What Google Analytics can't tell you (Part 2)

Engagement Time™

What’s wrong with “Time on Page”?
Last week, we looked at “Time on Page”, a statistic used by most traditional web analytics to gauge user interaction within a webpage. And while we weren’t Google-bashing (really, we weren’t!), we did point out some very serious inaccuracies with Google Analytics’ method of calculating “Time on Page”.

In this post we’ll be talking about “Engagement Time”, which measures exactly how long your visitors are actually interacting with your content. We’ll be looking closely at how it works, what it tells you, and how you can ultimately use it to improve your web content, conversion rates and ROI.

“Time on Page” vs “Engagement Time”
Traditionally, “Time on Page” tells you one thing and one thing only - how long a visitor has a web page open for. However, we’ve seen from countless observations that users will often open a new tab, minimize their browser or even go off and do something else while browsing a site. All of this is normal browsing behavior, but it gives rise to one major point: “Time on Page” tells you nothing about how long your visitors actually interact with your online content!

Time on Page can change drastically, as it can be skewed by normal browsing behavior Engagement Time however, provides steady, reliable and more representative statistic
Knowing how long your visitors spend interacting with each page is vital. You need to know what content keeps your visitors interested, what keeps them busy, and what bores them. For example:

•If you have a product features page, then you want to maximize your customer’s interaction with the page, keeping them interested until they convert.
•On the other hand, if you have a signup form, you want to keep the necessary user actions to a minimum, improving usability and minimizing form abandonment.

Knowing how long the average customer really takes to convert or fill out a form is the critical first step towards tightening the funnel and increasing your conversion rates.

That’s where “Engagement Time” comes in. ClickTale can tell you how long customers actually spend reading your content, looking at your pictures, watching your videos and browsing your products. Not just how long they had a page open for, but how long that page held their attention, and whether or not they liked what they saw.

istongtotong · 10-30-2009, 03:17 AM

Wow, nice information thanks I really need to know this.

quibusub · 10-30-2009, 03:40 AM

Glad to hear you're using this: I plan to keep it much more aggressively up-to-date than has been the case in the past, but don't hesitate to let me know if you find errors or need clarifications.
__________________
[LINKS REMOVED]

edosauyi · 10-30-2009, 04:24 AM

Google analytics is a nice tool though it cant be 100% accurate (just like every other tool) but it does some nice jobs.

cecille20 · 10-30-2009, 05:30 AM

I also notice it but I'm contented with Google Analytics service.

BTW for more information you can also use some tools like startcounter.

kevinkarl · 10-30-2009, 05:59 AM

Google Analytic is good Tools. Thanks For sharing the info.

lxh · 11-08-2009, 06:55 AM

Inaccurate yes, but not telling you anything about bounce rate, I have to disagree. First it can't be 100% accurate simply because its just a tool. Its not a human being that could differentiate the visitors who are actually reading from simply an open page. I can't actually imagine how exactly a tool could have a system to actually be accurate enough on that. Plus regarding the bounce rate, we at least somehow get an idea about it from analytics. =)