Comments on: Excellent Analytics Tip #11: Measure Effectiveness Of Your Web Pages

By: How much should you care about your home page?

How much should you care about your home page? — Tue, 19 Jan 2010 21:08:27 +0000

[...]
Some folks will tell you that your site’s home page doesn’t matter that much – Avinash Kaushik advises people not to “obsess” over it, and I have said things like that myself. Why? Because many of your customers probably enter somewhere else. But while it’s true you shouldn’t obsess, does that mean you should ignore your home page? Of course not.

In all likelihood, your home page isn’t the most visited page on your site. And there may be other deserving candidates for improvement. But, assuming you home page is one of your top pages, it’s easiest to think of it as just a different sort of landing page.
[...]

By: Kanika

Kanika — Thu, 17 Dec 2009 20:16:45 +0000

A quick thanks for putting up the MVT visual. I had understood it earlier also in one of your blogs, where you have explained the difference between AB testing & MVT. But, this visual made it clear. I would be glad if you could also post a visual of A/B testing too.

Thanks once again!
Kanika

By: Avinash Kaushik

Avinash Kaushik — Thu, 08 Oct 2009 17:08:40 +0000

Adrian: One effective way is to use the free onexit 4Q survey, users will directly tell you their intent.Another clump of ideas are in this post:Six Tips For Improving High Bounce / Low Conversion Web PagesIn your specific case I would recommend tip # 1, 3 and 6.Good luck!-Avinash.

By: Adrian Palacios

Adrian Palacios — Thu, 08 Oct 2009 15:44:41 +0000

Avinash,
I have a question about the best way to infer customer intent…not completely understand down to the last decimal place, but get a 'good feeling' of what they want to know.

Because the websites we oversee are heavily-Flash based, the site overlay in GA does not work. This is problematic because, as you alluded to above, a user's navigation can infer intent. Our sites are small, and don't have too many pages, but my theory is that if we segment traffic, certain groups will show an inclination towards certain pages. But I am not a magical statistician and so I am having a hard time wrapping my head around how to even see if there is "inclination" depending on source of traffic.

I think this could be important, because I want to tailor advertising messages to the intent that has already been shown. We can then see if changing the messages in appropriate places boost lead generation.

So 1) am I headed down the right path, or is this a waste of time?
2) if this is something worth investigating, how do I even go about it?

Thanks!!!

By: Bounce Rate: The Good and the Bad « Internet Marketing Blog with PaperStreet Web Design – Law Firm, Business & Professionals

Tue, 28 Jul 2009 18:34:29 +0000

[...] analytics specialist Avinash Kaushik has stated: “It is really hard to get a bounce rate under 20%, anything over 35% is cause for concern, 50% (above) is worrying.” With that in mind, let’s take a look at two examples of worrying and non-worrying bounce rates (i.e. higher than 50%):[...]

By: Avinash Kaushik

Avinash Kaushik — Fri, 17 Jul 2009 18:06:40 +0000

Carly: Here's a post on bounce rates:

http://www.kaushik.net/avinash/2007/08/standard-metrics-revisited-3-bounce-rate.html

You'll note towards the end the exception I make for blogs.

In a nut shell, measuring overall bounce rate for blogs might not be of value. I recommend measuring bounce rate on blogs for New Visitors because you don't want them to bounce, you want them to read your about page, you want them to sign up for your RSS feed, you want them to…. do something of value to you and all those actions mean a lower bounce rate.

-Avinash.