There seems to be similar sentiment on the web regarding the variations in statistics from one analytics software to another. Understanding how each system tracks clicks, impressions and pageviews would help to understand these differences. Sometimes users have javascript disabled or ad blockers running that keep some systems from tracking the analytics properly.  The stats on the

Bit.ly Link Counts

In regards to bit.ly and other url shortening services I found an interesting post regarding why these numbers might jump higher than what google analytics reports or your other analytics software running on your website.

“If you look at Tweetdeck and all the major Twitter clients they have a URL expander. Regardless of whether it is clicked on or not – the Twitter clients look up any new Shortened URL once at least. Hence if you do a URL shortener in tr.im or bitly – you can see on your analytics that within a few seconds you get hit several times…. thats the clients looking it up – Alas not a pair of real eyes!”  [original link]

That might answer the basic question of why external link trackers are different than google analytics but it doesn’t explain why inside bit.ly the counts can be different.

Based on the two images above you will see the Percy’s link shows with 3 clicks on the main link list inside bit.ly and then on the sidebar summary for top links it is showing the Percy’s link with 5 clicks. This data is on the same page and varies. My link totals are low in this case but for those with higher volumes this variation to grow even bigger.

Being a developer I understand it is possible that 2 different queries are being used to summarize the data or maybe from 2 different sources. Unfortunately, this should be a simple fix and be addressed quickly. I am currently very skeptical in the reported numbers when inside bit.ly the click numbers mismatch.

Twitter and Crawler impact

I found this interesting article to help shed some more light on the topic over at searchengineland.com.

Based only on referrers, at best, Google or any analytics program would have said Twitter sent 2 visits. But because I used tracking codes, I was able to overcome the lack of referring data and see that Twitter (itself or via applications or web sites using Twitter data) sent 9 visits. That means analytics packages might be undercounting Twitter visits by nearly 500%. [read original article]

Adwords and Affiliates

I am also working on some adword ppc campaigns and commission junction affiliate link programs. Within Commission Junction you can go from page to page and see variations in clicks and impressions. It seems one page uses up to the minute numbers and the other page uses a delayed value. Unfortunately this is not user friendly and gives me reason for concern.

One thing I have done to help at least with impressions and analytics for adsense and adwords is to link them with analytics. This should help tie the numbers together more accurately. However, this is only part of the overall internet marketing stats that an SEO needs to understand the data.

Getting better analytics

I will continue to look into the gap in values and see if I can come up with a better system to be confident in the performance of my own and my client’s websites and ppc campaigns.  I am still getting familiar with these tools and the best method for using them together.

If you have more information about how to build confidence in your analytics I would love to hear your solutions or explanations.

Related posts:

1. Twitter gets a url shortener
2. Kansas City SEO