Intention Clouds and the Standardization of Search


John Battelle and Matt McAlister both point to the release of the Intention Cloud -- an idea that combines tagging with querying Google Suggest in order to determine user intentions.

The Intention Cloud uses weighted lists to display the intents collected from databases of intention. The metric associated with each returned query, generally reflecting its popularity or traffic over a period of time, is used to calculate its printed size. Therefore, the more popular an intent the bigger it will be displayed inside the cloud. Colors are helping to intuitively separate the weights of the results.

With Google Suggest recently being added to Google News, are we seeing an attempt to standardize search terms? If so, does this minimize the value of long tail optimization? And is this better for information organization and helping people find what they're looking for?


Long way off, don't ya think?

If so, does this minimize the value of long tail optimization?

Google Suggest has been in the lab since '04. How long will the News version stay in the lab? If you asked 100 "Joe Surfer"-type people if they've ever heard of Google Suggest, how many would say "yes"? I'd peg that number at 5 or less ... in other words, long tail optimization seems pretty safe for the foreseeable future.

This first version of the

This first version of the Cloud is gathering its data from Google Suggest service. We are already planning new modules, to include other ‘database of intentions’, as Overture keyword research tool, Wordtracker, or anything that can be scrapped


another web2.0 business model?

"anything that can be scrapped[sic]"... so in other words, a service that can be stopped dead.

This sounds like that Tom

This sounds like that Tom Cruise movie (Minority Report?) where they had the 'Pre-Crimes' department that somehow looked into the future to stop crimes before they happened. I tried to watch it about 3 times but never got to the end.

This confirms my suspicion about Google databasing and tracking clicks on the "Remove result" button of its search results.

Looks Stupid

Intention clouds page looks like some keyword stuffed SEO page.

Waste of the 5 minutes it tool to load the stupid page.


I see no practical use for that cloud.

I think we'll have a dry season.

Not even an original idea

Its been done before -

Here's an impractical use...

I see no practical use for that cloud.

You can always use it to crash your browser. It did it to mine on the first try.


I think I had to answer to some comments here.
I am one of the guys working on the intention cloud, and I read your 'thoughts' with interest. First of all, thanks to the first poster who's asking a really interesting question. imho, such suggestion tools will never make their way to main SE's UI, for the reasons you pointed to. If users starts to pick up suggested queries, the diversity of the databases of intention will decrease. Moreover, a more down to earth reason for that is to be found in AdWords business.. Don't you think ?

Now, the rest...

@John Andrews

Yes, the scrapped services can be crashed easily (depends also on the quality of the scrapper..). That's why we'll integrate Overture, Wordtracker and other 'profesionnal' tools through their API. Maybe my sentence wasn't properly built and induced a misunderstanding.
Moreover, I must say that I'm a bit surprised when you ask the question "another web2.0 business model?"
Where do you see ANY business here?
Let me restate it again: WE DO NOT WANT TO DO BUSINESS. There is no ads on our site, there will never be, no services sold. See ? No commercial intent -> NO BUSINESS. So, business model? wtf? This is an experiment, and we plan on doing more in the same field. Stop thinking about money for a moment. thank you.


You do sound stupid. Any tag cloud looks like 'keyword stuffing'.. SEO is probably YOUR field, but it's definitely not what we're trying to achieve. And look at the god damn links, they're all pointing to google, not to a stupid adsense filled website.


Too bad :)
At least some are finding it intriguing, insightful etc.. I guess when we'll put online the version based on Overture, some in market research or SEM fields may find some concrete applications to it.
We'll also make a version with no concrete application at all, maybe even an audio tag cloud (yeah, why not ?). Just for experimenting with new mashups..


Yes, we know. We even linked you from our site.
But I don't really like your application of it. First of all, i find it very hard to read and the colors are so depressive.. I also find it quite hard to navigate in the clouds. Moreover, you don't really explain how you build them. We provide our source code under open source license. Finally, your site is filled with AdSense....
Anyway... we're not doing exactly the same thing. So, no bad feelings :)


Yup, we know it's crashing on IE6. We're fixing it this asap. It's not exactly the best browser to browse it anyway..



hmm.. interesting tool. I'm always interested by new vizualisation tools.

but, as far as I can see, Quintura seems to be closer to a tool like or other clustering+vizualisation tools.
maybe you should reread again the way we create the clouds. We're basing it on INTENTIONS: not what people gets in SERPs, but what they search for.

There are a bunch of tools similar to Quintura, and some way more advanced.
But it's definitely not similar to Intention Clouds..

Welcome to ThreadWatch

Hi az and welcome to ThreadWatch, a community of people involved with search and search marketing, and for whom every new page added to the web is a potential opportunity.

Maybe you don't like that people view your research as a potential opportunity, but hey, you published it. And you came to read the chat on ThreadWatch. As opportunities go, it seems to have achieved a "thumbs down" from the reviewers here at ThreadWatch. Don't let that bother you, though. It is a very specialized community.

As for web services like yours that rely on a third party to allow the scraping that provide's your site's value... well it's a dead business model. In your case, it's a dead website model. As soon as any one of those third parties decides to "play with your acess" your data is garbage. You won't know about the bias. Will you base your research on garbage data? Should we accept it sans critical review?

No wonder you think incrediBill sounds stupid. He has arguably the most advanced anti-scraper technology out there. What does that say?

What you might gain from ThreadWatch is a sense of perspective. Guys like incrediBill will use anticrawl technologies to block your kind of services, and blab about it in public celebration. Other people will will use incrediBill's technology to feed your service skewed data, as a means to influence for competitive advantage. Your sort of data (if not of questionable value) could be widely used by the keyword people, the trend trackers, the MySpace manipulators, and all sorts of "business people".

As for your "no commercial intent" puhleeeeze. Sorry..wrong forum. This is not a community of bloggers. We know better, even if you may not.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.