5 is still a bit limiting (Omniture allows up to 50), but overcomes a major blocker to using Google Analytics in the past. Interestingly, Google Analytics allows 3 different types of custom metrics: Page-level, Session-level and Visitor-level Custom Variables.

To test these out, I created a page name custom variable on each page. But more advanced implementations could include tracking by author, site section, logged in status, etc. Google Analytics just changed the game again with custom variables.

Ever since Techmeme released their Leader Board a couple weeks ago I was intrigued how sites like Computerworld could be ranked in the top 12, but only receive a few thousand clicks each month. Don’t get me wrong, I was very impressed/surprised/happy to see C-Dub even on the list (along with so many other traditional news outlets), but something didn’t seem right, as Carr points out, referral traffic from Techmeme barely registers compared to sites like StumbleUpon, Reddit, Digg and Slashdot.

So I started playing with numbers (don’t do this at home kids). Taking into account Techmeme’s “Presence %”, traffic data from Alexa/Compete/Quantcast and the volume of traffic my sites are receiving I’m getting a better picture of what is happening on Techmeme.com.

If my assumptions are correct, I’d estimate about 50% of Techmeme readers actually click through to the originating article. With 400k users, that creates about 200k clicks per month between 200+ Techmeme sources (technically, I don’t have a clue how many sources the site crawls, but 200 sounded like a decent number).

Based on the “Presence %”, the top 100 sites share 75% of those clicks (the top 5 sites garner 20% of total clicks). Even so, that’s still only an average of 8000 clicks per month to the top 5 sources, a drop in the hat for sites like the NY Times and Engadget. If you’re outside of the top 100, you’re lucky to get a couple hundred clicks.

So what gives? It’s pretty simple. Compared to Google News, Digg and Slashdot, Techmeme is tiny. Additionally, it seems Techmeme readers are content with just knowing what’s going on through headline/dek scanning.

Fortunately for Techmeme, the Leader Board seems to have bolstered traffic significantly. Clicks in October to CW from Techmeme have already passed August and September averages. If the Leader Board can supplant Technorati’s list as the default barometer of tech sites, Techmeme should become a larger blip on our referral radar.

Online analysis is leaps and bounds beyond anything you can get in any other advertising medium. The glaring problem is not a lack of measurement, but a lack of understanding about the types of measurement that can be performed online. Technology typically out-paces business, but I’m surprised in the multi-billion ad industry, that it isn’t closer.

It may be as simple as a lack of planning. Often times agencies, in an attempt to get their campaign to market as quickly as possible, overlook simple measurement techniques that would save them time and money when it comes to measuring campaign effectiveness. With a reported $2.5 billion being spent in online advertising last year, there seems to be a ripe opportunity for companies that can help ad agencies track their campaigns. Anyone interested?

UPDATE: Scott Karp always seems to say things better than me

As far as I can tell, they offer 2 main reports. The “recordings” mentioned above and horizontal heatmaps. Horizontal meaning they break your site up by time spent horizontally across the page vs. the actual areas that people spend time on, which is frustrating since you could have several important modules in the same horizontal area.

This is a neat tool for individual bloggers with low volumes of traffic, but large sites may want to take the following into
consideration prior to implementation:

* Every recording is of an individual visit. The largest package ClickTale offers is 9000 recordings for $99/month. If your site attracts more than 9000 visits each month you can set the tool to record randomly or until the 9000 recordings run out. My suggestion would be to identify sections of your site or specific templates that you would like to track and implement on that level. Then move the code to the next section once you’ve gathered enough data.
* I don’t pretend to be a page load expert so you’ll want to get one of your architects to examine the 2 separate javascript code snippets and load test them on your site prior to a full implementation.

All in all, ClickTale is fun to play with, but I’m not sure how much actionable data you could easily extract from it. But the relatively low entry cost makes it worth a look. Your designers will fall all over it.

1. Search result navigation is

1. iFrames – CSE loads in an iFrame which causes a slew of issues on its own:
1. You can’t track users beyond the first search result page. Subsequent pages are “hidden” to all tracking tools because they occur within the iFrame and on Google’s servers. Basically, this hides all user search activity on your site beyond the first page… no search effectiveness, no drop-off, no nothing.
2. Because the query is performed within an iFrame and on your site, the little pathing you could have done (from the first result page) is lost because once a user clicks on a result, they are taken to Google and then dumped back to your site–the path is lost and users appear to be coming from google.com/custom instead of your search results page.

Having been a Discover user for a couple months now, I hope they clear up some of the bugs with the current tool like slow loading graphs, random software crashes and the inability to break down domains to sub domains and URLs.

Features include multi-channel data integration, multiple breakdowns within classifications, 60 new custom events, menu customization, offline or delayed data collection, expanded advanced search, single sign-on support and several search engine updates. Additionally, long-time users will notice some tweaks to graph colors as well as some other minor design changes throughout the tool.

A few of these new features stand out. Multi-channel data integration will allow businesses to integrate external data sources into Omniture. This feature was available to larger clients previously, but is now seemingly included in the general release. Expanded advanced search is a welcome feature as it will allow you to easily save search parameters to get at data quicker. And the search engine updates help Omniture identify search engines to allocate traffic sources more accurately.

All-in-all, 13.5 is a good step forward from previous versions, but lacks major upgrades which is probably why they went with “13.5″ and not “14.”

I have 4 separate Google Analytics accounts with several sites in each, but so far only one account has been enabled on the new interface. Not coincidently, the activated account is my only one tied to an active AdWords account. Nonetheless, I have enough access to be able to review the new interface.

The tool looks to be built in Ajax and reeks of “Web 2.0.” Graphs are much more pleasant to look at and tools to interact/edit these graphs are intuitive. Data tables are very easy to read and most reports allow you to drill down further. For example, looking at referral traffic from search engines allows you to click on the engine to view which keywords/phrases made up that traffic.

The new tools available in GA include email/export reports (about time!), custom dashboards and trend/over-time graphs. While these are welcome, they were features that should have been in the first iteration of GA.

Final thoughts on the “new” Google Analytics are that cosmetic changes are very appealing and tool enhancements have caught GA up to some of the lower-tier analytics tools, but the new release is hardly worth switching for. GA continues to be a heavy Google AdWords user’s dream or a simple, free tool for the everyday blogger, but should not be considered for any enterprise-level analytics implementation on its own.

This raises 2 major questions for me. First, Google feed reader has only been available for about a year. In that time, according to these numbers, it has eclipsed the efforts of every other feed reader on the market. One of my feeds increased nearly 30% (about 20,000 more subscribers) on the day Google began reporting feed activity accounting for 23% of my overall subscribers. Is it possible that Google is over-reporting or that Feedburner is over-allocating activity to Google?

Secondly, and perhaps more importantly, if these numbers are accurate, Feedburner was under reporting activity to my feed by nearly 20,000 subscribers. That’s a significant misrepresentation of my feed traffic. This activity was obviously occurring prior to February 16th, why didn’t Feedburner role that traffic into an “Unknown” category? And, are there other sources of activity that Feedburner is not reporting because they cannot source it? If all of this is accurate, the RSS world may be a lot larger than anyone currently thinks.

Update: Rick Klau, Feedburner’s VP of Publisher Services addressed my concerns:

“We’ve always captured item views and clicks from [all] user agents… now we simply know how many Google readers there were to generate those views and clicks.”

“In order to track these requests differently, we have defined different event “types” via a new parameter. In this case, the parameter is WT.dl, with values to describe a pageview or an ajax request (or a mouseover, or an RSS feed view, or a “start” of a video…you get the idea). Within the analytics tool (WebTrends – no great surprise there ;-), we simply leverage the powerful analytics reporting (or new Marketing Lab Warehouse) to track these events independently (or together if needed). Pageviews remain as pageviews. Other requests are defined appropriately and tracked as needed to provide accurate reporting.”

Action orientated tracking is the future.? Page views are still important, but as web apps evolve, we need to rethink how we track elements on the page.