NYT’s list is the American list, based on Google.com. But what does the list look like in other countries? @skrubu looked at the suggestions that Google.fi gives and put together the list for Finland. I decided to do the same for Google.ca, and so here is the Canadian list:

A – Air Canada
B – Best Buy Canada
C – Canadian Tire
D – Dictionary.com
E – ebay
F – Facebook
G – Gmail
H – Hotmail
I – Ikea
J – Jobbank (government site)
K – Kijiji
L – London Drugs
M – Mapquest
N – NHL
O – OSAP (Ontario Student Assistance Program)
P – Plentyoffish.com
Q – quotes
R – Rogers
S – sports quotes
T – TD Canada Trust
U – Urban Dictionary
V – VIA Rail
W – The Weather Network
X – XSCargo
Y – YouTube
Z – Zellers

Unlike the Finnish list, the beginning of the Canadian alphabet bears a lot of similarity to the American list. But then things change: for I the first suggestion is Ikea, J is for Jobbank, a Canadian government site, and O is the Ontario Student Assistance Program. Interestingly, the only bank that makes it to the list is TD, and not RBC, Scotiabank, or CIBC. N is for NHL. And the online dating site Plenty Of Fish shows up for P instead of Pandora which appears on the American list.

I would be interested in seeing what the same list looks like in Spain, in the UK, in Germany, in Japan, in Russia. Anyone want to put those together?

Karri Ojanen

The idea of having the website to automatically adapt to each visitor and display slightly different pages for different users is, of course, not new. That’s what numerous sites, like Amazon, do by pulling information from user profiles, stored cookies, questionnaires and surveys. But the Sloan School system would adapt to unknown users within the first few clicks on the website by analyzing each user’s pattern of clicks. Websites running the Sloan system would detect a user’s cognitive style and morph to complement that style. The changes would be subtle and the users shouldn’t even realize that the website is personalized.

Well, of course. A site that would manage to adapt to every individual user’s needs would sell better – that seems like a no-brainer to me. And the changes should be so subtle that the user shouldn’t even realize the changes. But when I think of whether it’s possible to do that really, really well, I can’t help not to think of the infamous Office Assistant. Guessing what the user wants or needs next is hard in any case, even if the “guessing” is based on scientific formulas and research. I personally believe more in allowing users to do their own personalization rather than try to do it for them automatically. Emotionally I prefer a system that gives me the tools to make changes myself (if I choose so) instead of having the system try to automatically adapt for me, even if the changes are very subtle.

Nevertheless the Sloan School system sounds very interesting. In addition to guessing at users’ cognitive styles by analyzing their click patterns, the system would track data over time to see which versions of the website work most effectively for which cognitive styles. That’s what we try to do now by using site metrics and analytics, then manually study them and apply them to site revisions.

It’s not that I don’t think that the new interface can be better, and it certainly looks cooler than the rather bulky old UI, but I had just got so used to accessing the stats through the old interface and now in the new thing it feels like Google has moved a bit too many things around. The geo map overlay with the dots marking the cities/locations is one of my favorite things in the old design, and in the new version they’ve changed it to show numbers for continents, not individual cities, as default. I have to click through several views to get the details for cities.

I would think I’m not the only one who’s going to be lost for a little while after the old UI is gone. Good usability and design are, of course, important, but sometimes it seems like getting used to something matters even more than improvements in design.

Karri Ojanen

Tape Failure enables people to record user activity as they browse on a website. Like Google Analytics, Tape Failure requires no additional software to be downloaded. You can just paste a couple lines of Javascript into each page that must be tracked, and Tape Failure starts capturing the users’ mouse trail and clicks as they move throughout the site. In addition to recording videos, Tape Failure also collects “more traditional” type of statistics on the visitors.

After a quick look at Tape Failure’s statistics page, I think I wouldn’t use it alone to track users on a website, but if you combine Google Analytics and Tape Failure to monitor the same website, I think you’ll have a pretty comprehensive package.

Look at CMS Wire’s article on Tape Failure here.