At a minimum, I’m beginning to appreciate Twitter (and my new fave tool TweetDeck) as a better-than-delicious RSS feed. It is showing me what people that I personally know are smart and on-the-pulse of online activity are doing right now. Assuming they choose to share.

Case in point: Katherine Maynard (@KMaynard_SCC) tweeted about ClickZ’s article on measuring social media metrics: Social Media Sites Force Analytics Tools to Evolve. I was very excited to start reading the article because one of my main roles on the Army’s FOS project was to track and analyze all of the campaigns activity…not just log files and Google Analytics, but what we could learn from visitor activity on Facebook, YouTube, and Flickr. I was disappointed once again to learn that the samples they cited were either 1) just tracking referrers or PPC Facebook ads, or 2) built their own app to track visitor lifetime value/interaction over time. Another point that caught my attention was that Experience Project gets most of their referrals through organic search.

Learn more about the Army Faces of Strength Campaign.

We found in the FOS project that our external social media destinations were covering about 50% of the first page of Google for our key phrases. We also discovered that half of our campaign activity was occurring on the social media channels we set up and that the statistics available from each varied and had to be manually aggregated with the others to create a monthly analysis report that allowed us to create completely educated recommendations. I got pretty quick at it and produced some insightful reports, but I also know that we will be doing this for every client we can. And I don’t want to do it manually each time!

So starts my quest to automate the aggregation. Awhile back, I learned of tools/companies like ComScore, Quantcast, BlogPulse, Converseon, and BuzzMonitor but have never had the time to conduct research on them that wasn’t directly tied to a paying client project. Now, I have the opportunity to look into this periodically to see what is already out there without paying an arm and a leg, and what could be built. I know that we can use APIs like Facebooks to pull their Insight Statistics and Fan/Group data, but run into issues like Facebook not indexing their FQL page_fan.page_id field which means that it isn’t easy to pull data on your fans through a custom PHP application.

So anyway, I’ll keep you updated on where this goes and what I learn.

One of our clients was receiving a lot of 500 errors (5** errors are a variety of server errors) in their log files so we included an error trap so we would receive an email each time one occurred. Originally, it was setup because their XML file feeding their career search kept breaking and because it was coming from a third-party, we needed the heads-up. Now that that is fixed, we started getting some new ones.

These server errors were all happening on pages like this “/instmsg/aliases/*” where the asterisk was someone’s email alias. Turns out that when Microsoft Exchange opens an email sent by this site, it automatically looks to see if there is an instant messenger (IM) alias tied to the email address that sent it. Since Exchange isn’t on this server and there is no IM tied to the email address anyway, it resulted in an error. Thanks Microsoft.

No harm, no foul, but here are the details if you are interested: http://www.linkedin.com/answers/technology/information-technology/computers-software/TCH_ITS_CMP/164342-6857609

It is an ever-changing race to find statistics on online communication. While it is very helpful in deciding who is viewing what and what technologies you might want to use, the gathering of visitor data is not an exact science. Sure, it is better than using distribution number of a magazine, but there are so many variables, that you don’t always get them all. For example, Twitter publishes their user statistics, but doesn’t account for those that twitter using their cell phones or PDAs. This is probably a large percentage because of the short, text nature of Twitter communication.

That said, we are compiling useful statistics in a new category here for your viewing pleasure, our need to archive this stuff for reference, and because I love stats. Maybe it will be a widget someday.

Facebook

• More than 110 million active users (people who have accessed Facebook within the past month)
• Facebook is the 4th most-trafficked website in the world (comScore)
• Facebook is the most-trafficked social media site in the world (comScore)
• More than half of Facebook users are outside of college
• The fastest growing demographic is those 25 years old and older

LinkedIn

• More than 25 million experienced professionals from around the world,
• More than 150 industries listed on LinkedIn

Twitter

(Only includes website users. Cell phone and Twitterrific users not included)

• Total Users: 1+ million
• Total Active Users: 200,000 per week
• Total Twitter Messages: 3 million/day

Source: Facebook, LinkedIn, and Twitter respectively. Compiled by http://thebabyboomerentrepreneur.com/261/social-media-statistics/