I just noticed on the pMetrics blog that they are releasing an update soon that will fix the issue of only being able to select a single day as well as adding an API and logging of custom data.  This is great news and will constitute a HUGE improvement.  next on the list is adding export functionality and a more robust reporting interface in my opinion...but I'm excited to get the dynamic date range selector and API access!

I wrote a blog several weeks ago about some of the measurement tools available for measuring blog activity, and at the same time installed several tags from some of the providers in order to compare there relative performance. Now I will not try to compare the actual numbers since it's not only an exercise in futility since all tools use slightly different algorithms and business rules...but also because I don't care about that. I want to know which is the best tool currently available for measuring blog usage and for looking into the harder to find details such as inbound and outbound links, cross promotion, etc... So here are some brief results and opinions.

Measure Map - It's still not available, while I have been given assurances from a trusted and well informed source that it is not dead...but that's the only thing I've heard and their site is still unchanged. It's especially interesting since Jeff Veen who came over from Adaptive Path with the Measure Map purchase doesn't seem to be working on it either...at least the fourth paragraph of his blog post says he's not...

StandardStats - StandardStats and TotalStats are pretty much worthless in my opinion unless you post minimally to your blog and don't really want to look at web analytics. It seems to have problems if the "www" is omitted from the URL(thus creating two entries, one for www.myblog.com/content and one for myblog.com/content), so you have to do the math of combining them, which not only is complex since you can't export the data to excel, but it also muddies your topten list...which is the only display you get with the free version. Basically you should avoid this tool and use one of the other free versions.

pMetrics - While it has some great features as far as looking into the inbound and outbound links, intelligent content viewing unlike standard stats, and a host of other interesting tid-bits...it suffers greatly from poor usability and data manipulation. You can only view data one day at a time, and not only was that a terrible TV series...but it's a huge hindrance when you're trying to look at the activity on a certain blog post... One really cool option is looking at visitors on a Google map with full ajax integration, but google analytics has that too and it allows you to customize your date range.

Google Analytics - While it is decidely NOT a blog analytics tool, the new UI borrowed from Measure Map (did I mention that I want Measure Map?) is quite brilliant and easy to use. It provides a wealth of data but it still lacks the true blog measurability since it's orientation is more focused on standard websites.

So the end result is that we're still stuck using tools that don't fit the problem and no clear time line of when tools like Measure Map will be available.

I was recently reading a post about Google Website Optimizer and how they're changing the meaning of optimization. Anil Batra did a great job of looking at what optimization is and what it's commonly referred to, namely organic search optimization or SEO. He also explained multivariate testing as well and how it is being referred to as website optimization by google. Having a background in both SEO and multivariate testing I think Google is definitely creating unnecessary confusion amongst analysts who know SEO and multivariate testing...even though Google was probably trying to make it easy for people who don't know about either.

1. They're breaking existing nomenclature in the industry by naming a multivariate testing tool an optimizer.

2. While multivariate testing strives to optimize a certain conversion event by finding the optimal web page, it harly optimizes a whole website at once...unless you sell one product or have one objective and you have 1-2 pages in your site. Therefore using website in the name s equally erroneous.

I think it's better to stay within the accepted nomenclature since their product category is not new or challenging conventions, only challenging price points.

Google is clearly trying to expand their reach in the space by using a simple name since few people are familiar with multivariate testing. And I also assume there is some hidden agenda for google to dissuade customers from SEO since it makes more room for paid search...but I always assume google has an evil plot brewing.