Don't count on our help: What Google's suggestions can do to your message

Here is an interesting example of how Google's search suggestions can wreak havoc on your brand message - without you having any control over it, really. The examples in the case are in Finnish, but they are translated.
Nordea is the largest bank in the Nordics, and in Finland their catchphrase for many products such as loans is "Teemme sen mahdolliseksi" ("We make it possible" - the most boring and innocuous slogan a bank can have, to be sure).

Googling that with an English-language setting (used by many even in Finland) gets you the below suggestion: "Teemme se mahdollisesti" ("We might do it, possibly"), which you'll know to be a fair statement if you have ever negotiated loans, cards or accounts with them for that matter (especially if you are a small business).



And what can you do if you are the company in question? Do your research - in this case searches in relevant and related languages, misspellings and misunderstandings, and take findings into account in your PR, which you can use to organically steer the results in the right direction (I like calling in steering instead of optimization: there is so much snake oil sold in search engine optimization). Of course, in some cases a suggestion like this may only be triggered after you have composed your messaging, so review rounds are important as well.

Google's regression toward mediocrity

I have found that the user experience with Google has deteriorated over the past two years. In the past couple of months, it has worsened considerably.

First, Google started introducing more and more aggressive spell-checking and correction in the queries. Often, especially with obscure languages like Finnish, Google decides to correct the query, even if the correct spelling would yield numerous results. I remember either Brin or Page (or maybe it was Marissa?) saying that the ideal use case of Google would be that it only returned one result - the one that answers the user's query exactly. What happened to that notion? The "Did you mean"-function has been very useful and moreso with the advent of two panes of results, one for corrected spellings and one for sic - as it was spelled. Forcing repeated corrections, especially when it broadens the results set, can lead to regression towards the mean. With search results, can lead to mediocrity.

The more worrying new feature has in fact lead to a completely new use behaviour on Google. In the past, one could be fairly confident that the best possible results served to meet the query would be the top results, and rarely had to scroll to the bottom of the page. The new feature of omitting words from the query leads to the user having to scroll to the bottom of the page to make sure no words have been omitted. If there have been, they will discover a line of text: "Tip: These results do not include the word..." There the user is offered a link which leads to the query the user originally wanted to make. Here is an example, where the word "among" has been omitted by Google: http://www.google.com/search?hl=en&safe=off&q=aggression+among+diabetics (odd example, but relevant to research I was conducting recently).

This is incredibly contrived behaviour and arrogate development from Google. I completely understand that the feature has been extensively tested and proven to improve various problematic use cases, but it has also lead to an unforeseen difficulty in using Google. The strength of the search engine that I remember got me using it, was that it included all the typed words in the query, without having to resort to operators like '+' to force a word to appear. In fact, Google even told you that the '+' operator was unnecessary, since all the words were included by default.

There is very little difference in terms of actual experienced quality between the results of major search engines. Studies are divided as to whether there is perceived difference in quality (i.e. where the user knows the results have been generated by Yahoo! or Google), (Bailey, Thomas 2008).

I think Google has more to lose, and while it will be marginal, there are early abandoners as well as early adopters, and the margins will go first. Sheryl Sandberg had us read the Tipping Point as the first Google Book Club (yeah, we had one) book, and the simplistic learnings from there would be good to bear in mind when considering the importance of marginal users' marginal search results.

EDIT: just got a really good example where Google corrects my query making it useless:
http://www.google.com/search?hl=en&safe=off&q=atlas.ti+variable+playback+speed&btnG=Search&aq=f&oq=&aqi=

Tech> MacBook flaking... literally

Humm. A piece just came off my MacBook. When the black version came out, there were some rumours about paint peeling off and the like, but nothing alarming. I've only had the laptop for about 4-5 months now, but it's been used in all kinds of conditions.

The 4cm long piece came off where my right wrist rests and slides between the keyboard, trackpad and the the arrow keys. There's still a little crack there, so more might come off. I'd like to tape it, but maybe the extra ventilation will mean that the bloody thing will be a bit QUIETER (and I only have white gaffer tape on me...).

I'm on my way to Indonesia. Bye, Hong Kong. I think I'll put my Apple Care to good use when I get to Australia.

Tech> Stupidity> Copy-Yahoo-Paste-Google

Google promotes IE7. I'm quite ashamed that Google would do this. I don't even care about the competitive position - I only care about the fact that Google is promoting a program in a sub-standard line of programs, and one that is only available on the most miserable platform of all: Windows.

I don't know* what went wrong, but sure enough, looks like Google hosted and promoted a page with code copied from Yahoo!, and an image with the Y! logo just blurred off - not even substituted by a Google logo, but just blurred away. Miserable.

*as always if posting something about Google. See footer.