For Those Using Wordpress - What Stats Are You Using

[JB Lions]

For those using Wordpress, what stats are you using for site visitors.

Used to use StatPress Seolution but that's been discontinued, no longer available. Trying another person's version of it called NewStatPress.

Not including Google, just wondering what else people are using, what you would recommend. Thanks.

 

[Lawrey]

Check out statcounter

 

[abdussamad]

Server side stats aren't very accurate. Google Anayltics is the best.

 

[salmankaleem]

I've heard good things about StatCounter, Piwik, and Clicky. You might like Piwik as you can host it on your own server and provides analytics comparable to Google.

 

[lothos]

Server side stats aren't very accurate. Google Anayltics is the best.

actually, that's backwards. Server side stats are more reliable than remote stats like analytics.

 

[DU]

I just use Jetpack with Wordpress but then I haven't even logged on to my site in a few months I used it because it seemed easy and covered the basics that I was after in terms of site access.

That said, I think awstats and analytics would be good enough for me. I don't trust anything except Awstats

 

[zidane]

I agree with how Google Analytics is not as accurate as other tools. Tried them before and its evident that it doesn't work like it should. AWstats, clicky and stat counter are good decent options however I choose jetpack now. It works perfect for my wordpress websites.

Zidane

 

[abdussamad]

actually, that's backwards. Server side stats are more reliable than remote stats like analytics.

I must have missed this post of yours so sorry for the rather late reply.

Let me explain why server side stats are less accurate than javascript based ones. Server side stats can't filter out bots so they show inflated figures. Javascript based ones like Google analytics naturally filter out bots because bots don't execute javascript code. So if you want accurate stats you can't go wrong with Google analytics or statcounter.

 

[DU]

Server side stats aren't very accurate. Google Anayltics is the best.

Server side stats are the ONLY stats that are very accurate. You can slice and dice and run whatever algorithms you want to the data... you can also adjust pages to identify "bots" etc.

Server side does contain a lot of "noise" which you have to manually filter out. The accuracy you get depends on the effort you put in - but the server is THE most accurate. You can combine the data with your own logs for javascript execution/css poison links etc.

Google Analytics saves you a lot of that effort.

If it's THAT important to you would use both but for most people trying to gauge rough numbers either method would work.

Most people just want a rough idea.

 

[lothos]

I must have missed this post of yours so sorry for the rather late reply.

Let me explain why server side stats are less accurate than javascript based ones. Server side stats can't filter out bots so they show inflated figures. Javascript based ones like Google analytics naturally filter out bots because bots don't execute javascript code. So if you want accurate stats you can't go wrong with Google analytics or statcounter.

I don't know what stats program you're using, but my stats have no problems filtering out bots. Googlebot, MSNBot, MJ12bot, all sorts of bots get filtered out and not counted in my stats.

You do realize that some bots actually EXECUTE JAVASCRIPT, right?

I personally BLOCK google analytics, so please tell me how that would count my visit. I don't run their javascript, so I don't get counted. I WOULD get counted on a server side stats.

Server side stats are simply more accurate, period.

 

[flishess]

I use the wassup plugin to track what bots/spiders are doing on my site and Piwik for the rest.

 

[abdussamad]

I don't know what stats program you're using, but my stats have no problems filtering out bots. Googlebot, MSNBot, MJ12bot, all sorts of bots get filtered out and not counted in my stats.

You do realize that some bots actually EXECUTE JAVASCRIPT, right?

I personally BLOCK google analytics, so please tell me how that would count my visit. I don't run their javascript, so I don't get counted. I WOULD get counted on a server side stats.

Server side stats are simply more accurate, period.

Those are not the bots I was referring to. Search engine bots identify themselves as such in the user agent string. So they can be filtered out easily by any statistics software.

The bots that don't identify themselves are in the vast majority of visitors to any site that accepts user submitted content. That means just about any site that has a web form like a comments form, contact form or order form. I am of course referring to spam bots. These bots pretend to be legitimate users and can't be filtered out by server side stats. That is why you get inflated figures on server side statistics software and not on client side stat software.

The only bot that I know of that bothers to execute JS is google bot. As I pointed out google bot and other search engine bots are easily identified and filtered out by all analytics software.

In fact one very good way of stopping comment spam relies on the very fact that spam bots don't execute JS. It's called hashcash and there's a plugin for it for WP.

Finally people like you who block analytics are in the minority. Less than 0.1% of users. Most users are not very technical. If you've ever run a successful website you would know that. For example many visitors end up on a site by googling the URL! They don't even know how to use the address/location bar of their web browser.

 

[flishess]

For example many visitors end up on a site by googling the URL! They don't even know how to use the address/location bar of their web browser.

I've been trying to get my mother-n-law to stop doing this for years, she just doesn't get it.

 

[lothos]

Those are not the bots I was referring to. Search engine bots identify themselves as such in the user agent string. So they can be filtered out easily by any statistics software.

The bots that don't identify themselves are in the vast majority of visitors to any site that accepts user submitted content. That means just about any site that has a web form like a comments form, contact form or order form. I am of course referring to spam bots. These bots pretend to be legitimate users and can't be filtered out by server side stats. That is why you get inflated figures on server side statistics software and not on client side stat software.

I also said "all sorts of bots". Here is a list of bots that my server side stats filtered out just yesterday, including a bunch of UNKNOWN BOTS and SPAM BOTS.

Unknown robot (identified by 'bot*')
BaiDuSpider
Googlebot
Unknown robot (identified by 'crawl')
Unknown robot (identified by 'robot')
Yandex bot
MSNBot
Unknown robot (identified by empty user agent string)
Unknown robot (identified by '*bot')
Unknown robot (identified by hit on 'robots.txt')
BSpider
Unknown robot (identified by 'spider')
SeznamBot
Mail.ru bot
WordPress
Turn It In
Java (Often spam bot)
Nutch
MJ12bot
Sogou Spider
legs
ichiro
Alexa (IA Archiver)
FaceBook bot
MSNBot-media
Python-urllib
Exabot
Perl tool
Yahoo Slurp
BlogPulse ISSpider intelliseek.com
Phantom
NG 1.x
SurveyBot
Netcraft
WGet tools
larbin
Powermarks
W3C Validator
CFNetwork

As you can see, awstats and other server side stats most certainly DOES filter out all sorts of robots.

The only bot that I know of that bothers to execute JS is google bot. As I pointed out google bot and other search engine bots are easily identified and filtered out by all analytics software.

Then you haven't looked very hard. Here is an article from May 2008 (4 years old!) about bots executing javascript and screwing up analytics numbers.

Here is another article from June 2012 about bots showing up in analytics.

If you think your analytics are accurate because no bot executes javascript, you are mistaken. Lots of bots execute and parse javascript. I guarantee that you have bots being counted as visitors.

There's also a ton of services out there that run as "real browser instances" ( Keynote, Gomez, AlertSite, Pingdom, etc etc). These will also be counted by both analytics and awstats unless filtered out.

Finally people like you who block analytics are in the minority. Less than 0.1% of users. Most users are not very technical. If you've ever run a successful website you would know that. For example many visitors end up on a site by googling the URL! They don't even know how to use the address/location bar of their web browser.

Of course I realize this, but NoScript and AdBlockPlus (and similar scripts) are a lot more common than you think, it's more like 3% than 0.1%. Sites like mine that cater to a technical audience have a higher percentage of people blocking javascript which makes analytics even less accurate for me. I would rather have ALL the data, than MISS DATA because analytics is blocked or doesn't load for various reasons.

AWStats is based on log files and is intended to be a log analysis tool, while Google Analytics is intended to be a measure of website performance. Awstats doesn't analyze visitors as such, while analytics tries very hard to show you real visitor numbers. They have similar but slightly different uses and purposes. AWstats is meant more for system admins and is more technical. Google Analytics is meant more for Marketers and business analyts.

Google analytics will never be as accurate as awstats because of how it works (javascript on a third party server). It simply doesn't have ALL THE DATA.

Neither of them will be 100% accurate and both will have bots counted as visitors. If you're that anal about your stats, you really should USE BOTH or run an analytics type program with your server logs like piwik or mint.

 

[JB Lions]

Just an update, the NewStatPress I'm using is pretty good. Pretty much like StatPress Seolution. I usually used something like that and free stats like Extreme Tracking. It's public, which really isn't good, but I like that I can bookmark the url on my phone, so I can always check stats without having to login somewhere.

One of my new sites I have that with (only a few pages), I was getting Unique Visitors but it wasn't showing Referrer Tracking, I think mainly because within the last year or 2, Google default search is now secure, something like that. But checking today with NewStatPress, it showed me the keywords people were using in Google to find my site, so that's good it's getting some traffic.

Actually the searches are encrypted, started last year - http://searchengineland.com/google-to-begin-encrypting-searches-outbound-clicks-by-default-97435

"The change to SSL search also means that sites people visit after clicking on results at Google will no longer receive “referrer” data that reveals what those people searched for, except in the case of ads."

But as I said, looks like NewStatPress is picking those searches up, when Extreme Tracking isn't.

Trying the Statcounter for a few days, there's a Wordpress Plugin for it.

On my HostGator account, you have Awstats and Webalizer, just checked that. Good info but more than I really need.

On my Bluehost account, they have that as well but you have to enable it, where with HostGator it was enabled automatically.

The sites I do nowadays are very niche, so they're not going to get a lot of traffic and I just need something simple to look at what keywords are getting to my site.

 

[abdussamad]

I also said "all sorts of bots". Here is a list of bots that my server side stats filtered out just yesterday, including a bunch of UNKNOWN BOTS and SPAM BOTS.

All of those bots identified themselves as bots. The vast majority of spam bots don't do that.

Then you haven't looked very hard. Here is an article from May 2008 (4 years old!) about bots executing javascript and screwing up analytics numbers.

The very reason this article exists proves my point. If JS executing bots were common place there would be nothing special about them and no reason to write an article about it. This article shows that JS executing bots are in the minority and JS based client side stats are successful at filtering out the vast majority of bots.

Here is another article from June 2012 about bots showing up in analytics.

That article is about webmasters themselves setting up third party services to visit their site via automated web browsers. How many people actually do that? And those that do must know that it will screw up their statistics.

Of course I realize this, but NoScript and AdBlockPlus (and similar scripts) are a lot more common than you think, it's more like 3% than 0.1%.

AFAIK Adblockplus doesn't block analytics by default. NoScript blocks all JS by default and anyone using it is going to face all sorts of usability issues.

Spam bots are many times the number of legitimate visitors to a site. You should worry about them inflating your stats more than then 0.1% or 3% of your visitors who block analytics.

Also visitors who are blocking analytics are also likely blocking ads on your sites from loading. So in effect you can't monetize that traffic and it may be better that it doesn't show up in your stats at all.

 

[fm1234]

I use the native Awstats that comes with my hosting for tracking on basic stuff; once it's actually matured into a real site or starts doing business -- whether sales, lead gen or whatever other monetisation medium -- I install Piwik. In addition to being easily comparable (and in some ways superior) to Google Analytics, Piwik sits on my server, and as such I own the data, not a third party company. This is an important security consideration for me.


Frank

 

[lothos]

You said:

All of those bots identified themselves as bots. The vast majority of spam bots don't do that.

I clearly highlighted UNKNOWN BOTS that were filtered out by the script. They did not identify themselves.

You said:

Javascript based ones like Google analytics naturally filter out bots because bots don't execute javascript code.

I clearly showed you bots that execute javascript, and have since at least 2008. It's almost 2013, there are only going to be more and more bots that do javascript.

You said:

Finally people like you who block analytics are in the minority.

While it might be a minority, it is statistics that analytics misses. Analytics does not get all the data, so it cannot be more accurate than a server side stats program.

Also, once again, Sites like mine that cater to a technical audience have a higher percentage of people blocking javascript which makes analytics EVEN LESS ACCURATE FOR ME.

I also prefer to not hand over my data to a third party, especially a third party that has the power to determine how my websites get listed in their search engine.

I have refuted your every point and you've brought nothing new to the discussion, so I have better things to do than debate semantics here. If you have new data or statistics to show then I will entertain that, but otherwise I'm done with this discussion.

No stat program will be 100% accurate and If you're happy with your analytics, then enjoy it.

 

[iowadawg]

Have not tried this one, but will implement on two blogs of mine to see how it pans out.

http://wordpress.org/extend/plugins/official-statcounter-plugin-for-wordpress/

 

[JB Lions]

Trying another one out, one that was mentioned above, one I've always wanted to try but haven't to this point, Clicky, they have a WP plugin as well - http://wordpress.org/extend/plugins/clicky/

Put it on a site that gets some traffic and liking it so far, better than StatCounter.