Since I wasn’t in Chicago last week for SES it gave me an opportunity to catch up on my podcasts. I was listening to Shoemoney on Net Income and he had Neil Patel on talking about Crazy Egg.
I must admit I had the completely wrong idea about crazyegg when I first learned about. I thought it was really just a web analytics program. However that’s really not the case it’s a tool you should use in addition to your regular analytics. I feel the most powerful part is the ‘clickmap’ (see example). It looks like a heat map but instead of tracking eye movements it really tracks and displays ‘click density’ in visual form.
This tool can be invaluable when testing different layouts, or doing split testing on PPC landing pages. I’ve been using it a little over a week and have to say I’m pretty impressed. I’ll probably roll it onto this domain for a month or so and compare it to click tracks. Anybody else using crazyegg?
Popularity: 6% [?]
Related posts:
- Thanks to this Months Sponsors – February 2009 I’d
- Thanks to this Months Sponsors – April 2009 I’d
- Thanks to this Months Sponsors – May 2009 I’d
{ 12 comments }
I use it to improve my landing pages and through Crazyegg I am now making a few Euros more with my AdWords landing pages.
I haven’t had the time to try it, but it’s definitally on my list of to-do.
Question, does the heat map only show links that where clicked or is it using some sort of js that can show if the user tries to click something that is not a link. For example, a headline in blue that looks like a link but isn’t.
Also, does it have the same problem as urchin? If I have 2 links to the homepage, in urchin it can’t tell if I clicked the logo to go to the homepage or if I clcik the nav link. Can CE tell the difference?
Jeremy
Yep, I checked it out and love it. I’ve learned a lot about where people go from certain pages. It’s very useful for tracking how user follow the scent of information and therefore improving usability.
It does seem to register the actual link that was clicked – not just by the target URL.
WOW,
I just ran a quick test and this thing is AMAZING.
This is going to do amazing things for my arbitr… I mean PPC campaigns
Not only was it able to tell me which link was clicked even if multiple links on the same page point to the same url, it also is showing EXACTLY where on the link the person is clicking. I know if they are clicking on the word buy or the word of the product they are looking at even if it’s all the same link.
The only thing this is missing that I want, is to show me where people are clicking that isn’t a link. Not sure if that’s possible but that would rock.
Graywolf, thanks for the writeup. If you would like, email me and I can send you a special Crazy Egg account.
I tried them but they have some trouble with pages that use javascript so they can’t record the clicks. They say they are working on my problem.
Yes, I am using it. I used it to work out what was happening on our home page and realized how bad it was. Using the data I gleaned from this, I now have been able to convince everyone here that I major re-design is needed.
We have used CrazyEgg for a few tests and it is certainly a very cool product. We are quite anxious to make it work for us, since it really would provide insight that we could not get thought our normal click-tracking and AB testing.
However, we have had two problems. First, it doesn’t seem to be accurate in all cases. We have some very strong analytics in place and have validated their accuracy over the last 4 years (not the least of which are the CPC links by which we make a share of our revenues). In our test CrazyEgg under-reported clicks on these links by an order of magnitude. Other click counts were somewhat out of line with what we measure, and others were way out of line (and others were about right). It wouldn’t be so bad if everything were proportionally off, I guess, but I simply don’t trust the numbers.
We have communicated with them several times to see if there’s something about out site that might affect the results, but we have not gotten a resolution to date.
Second, in some testing, we realized that the entire page load seems to block on the completion of the CrazyEgg script. Recently we saw a number of cases where latency was very long, and under IE7 it appeared that the request never actually completed, which made the page appear to hang. The problem went away when we removed the script tags. It looks more serious than it is, but it gave our site the perception of being slow.
Finally, in a brief conversation with one of the CrazyEgg guys at the recent Vegas PubCon, he indicated that they were changing their orientation away from usability and heatmaps kinds of things to AdSense optimizations, and so on (claiming a lot of competition, which I have not yet found). I didn’t quite understand this, but came away feeling somewhat uncertain about where they are planning to go.
So despite this underwhelming endorsement, we are very hopeful that having made lots of money during this merry holiday season, we can now afford the risk of trying CrazyEgg again. Their value proposition for us is very strong, but only if we can trust the data, and trust the service to not interfere with our site’s normal operation.
Tom
The big plus with Crazy Egg is that you can see the actual location of the click. In my brief use, I determined:
* Visitors were clicking on a graphic that was not linked, which I am sure frustrated them. It is easy to make it clickable, but we had not thought of that before.
* There were three words on a button at one of our main conversion points. The heatmap showed us the actual location of the clicks, which was over two of the words. We, of course, dropped the non-performing third word and made the two performers a little larger.
The above would never have been known if we had only used regular analytics. So, I agree that CrazyEgg is great to be used along side existing analytics.
The testing we have done recently was not really very serious as we are going to have a completely new site within months, but thanks to CrazyEgg, I have specifically told my (outsourced) designer that the final revisions on the design will not occur until a month *after* the site has gone live with Crazy Egg code at key locations so that we can see how real users interact with the design.
Cool tip, I’ll check out Crazy Egg. I just installed Feedburner stats on my blog, and I’m already running google analytics. I’m not sure how many javascript files I want to have to excecuting on every single one of my pages. There website and heat maps definitiley look cool though.
I agree whith TOm.
Crazyegg is a cool Tool but the values (counted clicks) sometimes are totaly wrong.
That makes it difficult to use this tool for regular testing, as values are different from test to test.
Comments on this entry are closed.