Filter visitors by page, better segmentation - and more!

We just released an update that's been two months in the making. On the surface, there are only a couple of major differences you'll notice, but they are very cool ones for sure. What took so long was designing and implementing an entirely new system for how we store page URLs, and converting all of the existing data into that new format (many hundreds of millions of database rows). This fixed a couple of annoying problems we've had with the system from the beginning, and let us build some awesome new features on top. Let's get to it.

Filter visitors by page!

You can now filter your visitors by a page they have viewed. This has been a big request. Works in the API too, just use the parameter "href" with the "visitors-list" type.. You should only be sending the path and query (e.g. everything AFTER ""), unless the page you want to filter by is on a mirror or sub-domain, or it's an outbound link. For example here's how we'd grab all visitors who saw our /users/ page, and all visitors who clicked our company's web site in our footer (

New visitor segmentation!

When you filter your visitors by anything, the only type of summary data you used to see was basically everything from "the basics" dashboard module, but applied to these visitors. But now you can see anything you want about them - top searches, top countries, etc. This is particularly useful when filtering by page. Now you can see the top searches hitting any page on your site for example. We plan to add on to this quite a bit this coming week, so for now it's slightly rudimentary, but it performs its purpose well. Click on the red "View these visitors..." link (shown below) to choose what type of data you want to see. This feature is for Pro members only.

Links so hot you'll want to eat them!

Links to external content now hide the protocol by default, which saves precious space (particularly on the dashboard). We've also changed it so it highlights the domain name and the rest is slightly faded. This is inspired by the Google Chrome address bar - I really like how easy it is to pick out the domain when looking through links now.

Also, when viewing your visitors list, there is now an icon next to the referrer to indicate which type of referrer it is (search, social media, etc).

Links to external content have changed!

Used to be when you viewed your "links" or your "content", clicking on an item would take you directly to the external page. That's useful but inconsistent with how the rest of the site works, where clicking on anything else (search, country, etc) would instead filter your visitors by that item. Links and content have been changed to match this behavior in all areas of the site. Now, anytime there is an external link, there will be this arrow next to it. Click that arrow to be taken to the external link. You can see this in the screenshot above.

Much better support for sub-domains and mirrors!

If you had multiple domains pointing to the same site, or had the same code on a bunch of sub-domains for one site, you couuld still track that fine, but the "content" we tracked was never linked to the right domain that it was actually viewed on. Now, we'll store the domain it was viewed on, if it's anything other than what you have in your "hostname" field in your site prefs. Example:

No more dupes!

The new pages system gets rid of the dupe problem for your content. Unique pages used to be determines by a combination of the URL and the title. This resulted in a lot of duplicate entries for the same page, because of reasons like the visitor having Javascript disabled (no title), the visitor using an old browser sending us the title formatted slightly differently, etc. This will happen no more, although it will only affect your stats from this point forward.

In conclusion!

That covers all the new features that are obvious on the surfage. We hope you love them like we do, but as always leave your feedback, good or bad, below.

Also, there are a few gotchas that might catch you off guard. Let me cover that. First, pages will not have any titles until they get a page view under the new system. This is related to getting rid of the dupes problem. Give it a few days and almost all of your content will have titles again. Obviously your popular content should already.

Second, filtering by page may not work right away for all of your pages, because there were so many duplicate IDs. If "your" ID for a URL was not the "first" ID for that URL in our system (on your database server), then they don't match up right in the database. So filtering by a page URL may only work for some of your pages from "this point forward" (as I seem so fond of saying). I do have an idea of how to fix this so filtering will work for all pages, for any date, but it'll have to wait until tomorrow so I can test it.

Last, the new segmentation drop down menu doesn't work right in MSIE. I could not figure out the problem before release, but I will try my hardest to get it fixed on Monday.

Ok, now go enjoy the new features!
25 comments |   Feb 22 2009 7:46pm

What are your top 3 feature requests for Clicky?

We'd love to hear your top 3 feature requests for Clicky. This service has been in non-stop development for almost two and a half years, and there are still a billion things we want to do. Our todo list is 10 miles long, so it's hard to decide what to do next sometimes. That's where you come in.

Excluding all the stuff that's coming on our March roadmap - which includes things like filtering visitors by page URL, per-page analysis, and super mega ultra spy - what are you top 3 feature requests?

UPDATE This isn't a poll because we don't want to restrict your choices. We want to know exactly what you want, even if it's absolutely ridiculous.
59 comments |   Feb 18 2009 8:11pm

Feedburner has been fixed

All of you who have moved to the new Feedburner "platform" have been noticing that Clicky can no longer access your Feedburner data. I'd like to rant about the embarrassingly awful job Google did with this migration, but I'll save my efforts for something more useful. What matters is that it should be fixed if you are on the new platform. If you continue to have problems, please contact us or leave a comment here - but it should be working for everyone now!

UPDATE One of our users asked about why their FB data from more than a few months ago (around when they migrated) had disappeared. I just want to clarify this is a side effect of your migration. When you move from the old Feedburner to the new one, your data doesn't carry over - you start fresh, as if you had just signed up. I mean, that would just way too convenient, wouldn't it, to have your data carry over? Surely you weren't expecting them to spend the 15 minutes it might take to write a script that exported your data from the old database and imported it into the new one?
22 comments |   Feb 16 2009 4:13pm

Lots of new tracking features, including OnClick goals and internal script logging

Lots of new stuff related to tracking being released today.

OnClick goals

You can now declare goals from Javascript events, aka "OnClick". This has been a big request. Docs.

Internal script logging

We have documented the API that our tracking code uses to log data, which means we officially support this now. You can log any data you want to it, anytime you want, from anywhere. Validation is done with a new sitekey, called the "Admin sitekey". This is available in your site preferences. We know this will be insanely useful to all of you who use internal redirects for outbound/affiliate links. I get questions on that all the time, so you should love this! We also wrote this PHP function as a "simple" way to add this functionality to your own app. Use something other than PHP? Rewrite and send it to us, we'd love to offer it to our users.

I'm pretty sure that Clicky is the only service in the world that offers this type of functionality. Please tell me otherwise if this is incorrect.

Better customized tracking

There is now just a single Javascript variable called "clicky_custom" to deal with all aspects of customized tracking (adding custom user data, changing the pause timer, etc). Much cleaner and easier to work with, and a few extra features too. "pageview_disable" to disable the automatic logging of a page view on initial load, and "href" and "title" properties to customize the URL and page title that is logged for that page view. The docs have rewritten from scratch with all the new stuff added in. Backwards compatibility is built in so you don't need to change any of the existing features you're already using, such as custom data tracking. To use the new features, you'll need to update to the new methods though.

Update sessions after-the-fact

Using the same API mentioned above, you can update sessions well after they are expired, to add your own custom data to them, and declare goals and/or revenue. I get many emails from people who can't add tracking code and/or javascript variables to their shopping cart pages. Now you can update sessions later if you want, to declare the goal and revenue for that visitor.

It's 3am. Time for bed.
8 comments |   Feb 13 2009 2:42am

Google's new Ajax-powered search results breaks search keyword tracking for everyone

Do you run a web site? Do you use a service such as Clicky, Statcounter, AWStats, etc, to monitor the traffic to your web site? Do you find it useful how these services can tell you what search terms are leading visitors to your site? Have you ever wondered exactly how they are able to do that?

No, you probably haven't. But that's ok. Just know this: a major update that Google is testing has completely broken the ability for any external analytics service like Clicky to determine the search query used by a visitor arriving at your web site. Why would they do such a thing? Who knows. They aren't talking.

Knowing the search terms that are driving traffic to your web site is probably the #1 reason to use any kind of traffic monitoring service. And Google owns a huge chunk of the search market. If this update goes live for everyone, it effectively means that 2/3 of all searches leading to the average web site will be a complete mystery. This is huge.

So what is the problem exactly? Normally when do you a search on Google or any other search engine, the search term used become part of the URL. A search for Clicky, for example, gives you this:

When someone clicks a search result on that page, that URL above is sent as the "referrer" to the target site. An analytics app running on the target site can parse the referrer string and extract the word "clicky", and store that as a search that occured for that site. This is obviously very useful.

Here's what the new search result URLs look like with the new "Ajax" feature:

See how there's a hash mark # in there now, and the "q=test" is after it? The problem is that web browsers don't send anything after the # in the referrer string. This means organic searches from Google will now show up as just "", with no search parameters. In other words, no analytics app can track these searches anymore. I started noticing lots of hits from just "" recently in our own search logs. I thought maybe it was just a bug with Clicky. But then one of our users contacted me about this article, and my jaw about broke from hitting the floor so hard.

This change isn't live for everyone yet so if you go to and do a search, you may not see it in action. But it is happening for me when I go there and do a search.

So what can we do about it? If you run a blog, write about this. Submit this story or your own story to large tech blogs like TechCrunch, CenterNetworks, ReadWriteWeb, GigaOm, etc - no large site has written about this yet, and one of them needs to. Post in Google's Web Search forums (there's no way to directly contact them about web search unfortunately). Do anything you can to spread the word and let Google know that this is not acceptable.
80 comments |   Feb 03 2009 9:50am

Next Page »

Copyright © 2019, Roxr Software Ltd     Blog home   |   Clicky home   |   RSS