Comparing Web Statistics Services - Sitemeter, Go-Stats, Extreme-Tracker, 103Bees
At the weekend I put several hit counters with full public access on The Wardman Wire blog for comparison. This post is a comparison of the results so far. My first posting “Sitemeter: Alternatives for Refugees” explains why I am comparing different online traffic metrics services.
Caveats
- At present this is a low volume (but high quality
) blog, as it is very new. Therefore I have only drawn crude conclusions. - I have had some problems setting the time zones to make sure I am comparing the same 24 hours - which is another reason to only draw broad conclusions.
Comparisons
I’m comparing 2 figures from each service:
- Unique visitors: how many ip addresses visited the site.
- Page views: how many web pages were viewed by those visitors.
Visitors Monday midnight to Tuesday midnight
| Source | Unique Visitors | Page views | Conclusions |
| Statcounter | 64 | 96 | |
| Go Stats | 59 | 92 | |
| Extreme Tracking | 36 | 62 | Appears to miss a lot of traffic |
| 103 Bees | 36 | not available | Appears to miss a lot of traffic |
| Server Stats | 122 visits | All services appear to miss quite a lot of traffic! | |
The Verdict so far
From these figures, I’d say that Extreme Tracking and 103Bees could be excluded from consideration. I want more data to confirm that finding, however. I would be really interested to hear results from someone who has more traffic.
I have decided to leave the exercise running for a full week. Return on Monday 15th April for more reliable results.
All the stats are publically accessible, except for 103Bees where that visibility is not available. The icons are all at the bottom of the sidebar (but only for the duration of the exercise).
What are your experiences with your alternatives to Site Meter?
Other comment
From Doctor Vee and Net Wizard.
Tags: Site Meter, 103bees, gostats, go-stats, extreme-tracker, alternatives to sitemeter, sitemeter serving tird party cookies, tracking cookies, leaving sitemeter, comparing statistics, statistics services, web counter, blog stats
[tags]blog stats, Site Meter, 103bees, gostats, go-stats, extreme-tracker, alternatives to sitemeter, sitemeter serving tird party cookies, tracking cookies, leaving sitemeter, comparing statistics, statistics services, web counter[/tags]
Article Series - Blog Statistics: Sitemeter Alternatives
- Site Meter: Alternatives for Refugees
- Comparing Web Statistics Services - Sitemeter, Go-Stats, Extreme-Tracker, 103Bees
GoStats.com allows you to block your own PC from counting; you may miss a few of your own hits this way.
-Another thing to note is the order of the counters on the page: if a person with a slow net connection is clicking quickly through your site, you may miss a hit or two most likely for the trackers lower down on your page. (however that doesn’t explain the significant difference with the lower ones)
-Server stats will always show more as it will include bots, image theifs, pings (or other non browser behavour)
Matt - The Go Stats figures for yesterday (Wesnesday) were 138 visits/161 pages vs figures for my server stats (using the SlimStat plugin) of 164 visits - which are far closer than the figures in the report.
Richard
Thanks very much for your comments - I really appreciate the input.
As I said, these are tentative conclusions.
I’ll be doing some more detailed comments at the end of a full week, when I should have a better set of data - something like 1000-1500 visits. This should also deal with any “time zone” problems.
My experience on filtering webstats for larger (work) websites, is that - foe example - up to 25% of traffic on a site with 250,000 page views a month was bots of various sorts (including watch dogs). This was filtering against a list of several thousand ip addresses used by search engines etc.
I’ll be doing a more detailed comparison of the services offered at that point, and crunching my server logs in some detail - and inviting comment from the services.
I need to get around to banning image and video / audio leechers.
Once more - thanks for dropping in.
Matt