Facebook Pixel
Join our Facebook Community

The SiteMeter Frenzy: Is it Really Necessary?

This post was contributed by Aaron Brazell, a regular ProBlogger contributor.

Last year, in the early days of my “problogging”, I was in New York City with a friend of mine who was attending Ad:Tech. He was attending, I was working on a book. Later in the evening, he was telling me about some of the encounters he had at the conference. One of the encounters struck me as odd.

The advertiser refused to deal with anyone who didn’t publically display their stats based on Sitemeter. He refused to think of any other stat tracking as accurate, even though javascript-based stat programs are simply inaccurate by default.

As time has gone on, I’ve noticed a frenzy among the self-named top echelon of bloggers to display Sitemeter stats in some publically accessible way. It’s almost a mantra in some circles. I’m not sure where this trend that says respectable blogs must use SiteMeter got started, but WebLogsInc made it popular – and where they lead, hordes of others follow.

But is it really necessary? My answer: No, not at all.

Jeremy Wright has this thought about SiteMeter:

There’s two sides to the SiteMeter coin. First, as a blogger, if you have no other stats it’s better than nothing. Personally, I wouldn’t display those stats, though, because I don’t believe SiteMeter has “integrity” in its statistics.

It’s nothing wrong with SiteMeter per se, it’s just the reality of using non-server based stats solutions: they simply will never be accurate. I know for most bloggers a server-based solution isn’t possible since they’re using Blogger’s and TypePad’s and they really just want more detail than those services provide. In which case, rock the SiteMeter (or Performancing Metrics)! Just don’t think it’s the be all and end all of statistics.

Stats are important. They are VERY important. Are they necessary for a blogger? Yes. Are they necessary for an advertiser? Absolutely. Are they necessary for the general public? Probably not.

Should choosing not to display public stats alter the net effect, net reach and net approach of the blog? Absolutely, positively not.

If I’m a reader, I read because of content. I don’t care what the blogger is making in terms of money, if any at all. I don’t care what kind of traffic the blog is getting. If I like what I read, I’ll read some more.

Tangent: At b5media, for instance, we don’t talk stats specifically. It’s our policy. it’s how we operate. We may talk generally, but we offer no specifics? Why? Because to you, the reader, it doesn’t make one bit of difference to your reading experience. Advertising deals are handled behind closed doors with the advertisers, mano y mano. There’s no need to air “the business” in a public forum.

So what is the best means to collect stats?

Personally, at Technosailor, I use 6 different stats packages. Each give me a different view. None of them are 100% accurate, however, server log based packages do tend to provide the most accurate information as they gather data that javascript-based programs simple cannot.

  • AWStats – AWStats is my bread and butter. It provides the best overall analysis on my traffic but has the shortfall of only being updated once a day.
  • Performancing Metrics – Pmet is one of my favorite tools for analyzing blog-specific qualities such as comments, post views, etc. It is geared for the blogger so it works well for what I do. However, it is javascript based and so it tends to be “low” on the analysis accuracy front.
  • SiteMeter – I use this program the least, checking it maybe once a week. The stats are not public and never will be. This is also a javascript-based app and tends to be on the low-side.
  • Feedburner – Feedburner stats are the only stats I publically display and I do it more as a motivating tool for me. I see the numbers and if they dip, I need to do better. Feedburner (the Pro package) gives great insight on where my feed is being used, and how well my RSS subscriber base is doing.
  • Google Analytics – I don’t use this often, especially since I’ve removed most of my Adsense, but it does offer another perspective on performance.
  • W3Counter – Very nice package that I have no idea if it is open for new members. I helped the developer by providing my Feedburner information for him to use in development. This package tracks both websites and feeds. Very slick, but a bit low as it too is javascript based.

Bottom line, no single stat package will give a completely accurate assessment. It’s usually best to try to use a handful of tools that each major in different areas to get an accurate assessment of what your traffic looks like, and none of that information is necessary for the public to enjoy your blog.

About Guest Blogger
This post was written by a guest contributor. Please see their details in the post above.
Comments
  1. That’s what I call geeky, Aaron – 6 stats packs! I don’t think I’d sleep if I checked that many.

    Question: since Shortstats doesn’t work on WP 2.0x, which single plugin package would you recommend for WP 2?

  2. W3Counter is indeed open to new members. Feel free to try it out :)

  3. I have jumped on the Feedburner bandwagon with a site that I just launched called OneMillionSubscribers.com basically I am trying to subscribe 1 million people to my RSS feed. Check out the site. Why am I doing it? Just to see if it can be done.

  4. You’re right… readers don’t care about stats. Publishers care, like when one of my posts got dugg and slashdotted yesterday, but that doesn’t matter to the reader, and there’s no reason to publicly display stats just for the sake of displaying stats.

    I ended up using Mint (http://www.haveamint.com) for stats on my two blogs.

  5. AXS is a good choice, though development has stopped:

    http://www.xav.com/scripts/axs/

  6. I am hosting my blog on a linux server, so I’ve set up AWSTATS to refresh on an hourly basis by adding a line to crontab to run the following script at 15 minutes after the hour, every hour.

    perl /usr/local/awstats/wwwroot/cgi-bin/awstats.pl -config=www.mckeay.net -output > /var/www/html/awstats.html

    This works well from me and keeps my obsessive/compulsive stats checking satisfied. You can look at my stats at http://www.mckeay.net/awstats.html if you care to.

  7. As I understand it, the problem with some of the non-java-based solutions (such as awstats) is that they count bots as visitors. Correct? In which case, just because it’s higher doesn’t mean it’s more accurate (at least not in terms of counting real, live readers who stop by your site).

    My experience with Mint has been that its visitor count is even more conservative than that of SiteMeter.

  8. well, I used Webalizer and it provided me with extremely innacurate stats :(
    Now, I’m with Google Analytics and Sitemeter.

  9. How bizarre that some should think it’s a sign of being a pro blogger to publicly display stats – I actually took SiteMeter off my blogs after a period of time. Oh well, each to their own.

    I use AWStats which suits me – I used to use Webalizer or Wusage which I liked better but can’t be bothered to try and install either of them on my servers at present.

    Stats are important, not so much for advertisers as far as I am concerned (yet) but for motivation, checking trends, referrals etc. Having reliable stats is essential.

  10. @ Patrick Grote:

    Until a few months ago I had AXS installed on all my sites as well. It was like having a log analyzer (like Awstats) without the horrible skewing from spiders, leechers and hotlinkers inflating all the numbers. I liked how I could follow a visitor through my site one page at a time. That’s one of the features I made sure to integrate into W3Counter.

  11. I have Google analytics installed on a couple of my sites. I find the stats they provide are good over the long term. There is just sooo much information provided to easily glance at. If you are looking for daily stats Sitemeter seems to work well in my opinion.

  12. Like you, I use AWStats because of the detailed analysis it provides. The fact that all of the information is obtained directly from the raw logs, it provides the best picture.

    Depending on your server setup, you should be able to run multiple cron jobs to have it update more than once per day.

  13. Good post. Thanks!

    What Feedurner stats do you display publicly. Just number of subscribers? Are there options to display more than that?

  14. Just number of subscribers in a chicklet. However, it’s not there at the moment as I changed themes yesterday.

  15. Sitemeter stats are the “official” stats of the TTLB Ecosystem.

  16. Question: since Shortstats doesn’t work on WP 2.0x, which single plugin package would you recommend for WP 2?

    John: I don’t actually like any of the plugin based stats packages. I don’t want anything that is going to make my database grow everytime I get a visitor. I am hosting a friend’s blog and yesterday we ran into an issue and had to import his database into a new database. Long story short, he has a tiny blog as blogs go but his database was 287MB! That’s uncalled for, unneeded and it was all because of a WordPress stat plugin that logged everything into MySQL.

  17. I just signed up for W3Counter and so far, I like it. Thanks for pointing it out.

  18. Not to mention, that Sitemeter javascript can be s-l-o-w. Agree fully, there’s no way I’d make my stats public; noone who doesn’t want to contact me directly needs to know.

  19. I use Sitemeter myself. It may not be 100 percent accurate, but its enough for me.I don’t see the need for all this controversy. Mine is public, but I only set it like that so I can access my stats on my blog’s page instead of signing in.

  20. Leon: By all means, use SiteMeter anyway you want to. The point of the entry was that some entities make it seem like not displaying publically is a mortal sin and offers less credibility to the blog.

  21. I could only see adding a public site meter to my blog if I was actively recruiting advertisers. That would be an easy way for them to get a feel of the popularity of the blog before e-mailing me for more details.

  22. I check both Google Analytics and sitemeter and i am a big sitemeter fan. I agree with you that most readers (99%) do not bother about sitemeter stats.

  23. I always thought that JavaScript-based stats packages were more accurate as they only counted your human visitors. What am I missing?

  24. Purely JS-based stats like Google Analytics won’t count users with javascript disabled, or viewing your site through a cell phone without javascript support. It’s probably only 2% of your visitors, but I suppose that’s underreporting…

    W3Counter throws in a portion on its tracking code with an image-based tracker to pick up those users, and provide analysis of JS support in various browsers for your site.

  25. My site has both Webalizer and AWstats. I know there are bots being counted on webalizer but there seems to be a large discrepancy between visitors, pages and hits. For example if I take, for the first part of June, Webalizer hits minus bots (as reported by AWstats) there is still a 35% difference. The other day I had 3,500 pages registered by Webalizer and 1080 by AWstats. And my AWstats unique visitor stats (500-700 a day) are half my Webalizer stats. Are there really that many bots out there or is this comment and trackback spam? Is Webalizer really overreporting that much or is AWstats under reporting? Are there any relaible resources that cover this?

  26. I use sitemeter for my blogs. I like it. It is simple tells me where my visitors are coming from.

  27. I’m using webstats4u.com. Previously is alright, after that the new 3.0 version, it starts to have that irritating random popup.

  28. I use AWStats, and I think it’s great. It’s enough for our website, although I’m sure some might require more detailed stat information.

  29. […] I’ve never really been one for posting statistics on my web sites. Like Aaron says, it’s not really something that a lot of people that are reading my blogs care about. […]

  30. I use sitemeter! i cant mess around with 6 statmeter too much for me

  31. I just switched to Sitemeter the other day, formerly using Nedstat tracker. But since Nedstat switched to Webstats4U, some users complained they were being served pop-ups on my blog. So I removed it and went with Sitemeter.

    I just have a personal blog, don’t do much traffic, just friends and family and people who know me (personally or over the web). But my wife and friends like the map function, showing the last several locations of visits. They think it’s neat a little blog from an Indiana guy has people from other countries that visit. Makes the world a smaller place to them, and that can only be a good thing, I figure.

    I haven’t used Sitemeter long enough to comment on it’s accuracy. But the price is definetly right (free!), for the functionality it offers.

  32. I DONOT like WebAlizer.
    Main reasons:
    1) Not very correct stats
    2) Refspam through webalizer logs

    Refspam is popular in my country, and in case they make it more often the site with WebAlizer may me ddosed.

    Thats what i think

A Practical Podcast… to Help You Build a Better Blog

The ProBlogger Podcast

A Practical Podcast…

Close
Open