blog.collarity.com

Collarity has been chosen as one of the hottest one-hundred private companies by AlwaysOn. Earlier today, AlwaysOn released their fifth annual AO100 list of companies representing the most promising entrepreneurial opportunities and investments in the global technology industry.

Collarity and other AO100 winners will be honored at the upcoming AlwaysOn Stanford Summit, being held at Stanford University from July 31st to August 2nd. Levy Cohen, Collarity CEO and Founder, will be presenting an overview of Collarity’s community-driven site search and content discovery services for websites at the Summit on August 1st, at 11:15am PST.

A recent New York Times article titled “The Human Touch That May Loosen Google’s Grip” has sparked some fresh discussions about adding more people-power into the search relevance equation — a subject near and dear to our collective hearts here at Collarity. Primarily a reaction to the entry of Mahalo, a new search engine site that enlists the aid of human “Guides”, the article points to the trend of balancing machine-based algorithms with “editorialized feedback”.

Google’s Matt Cutts was interviewed in the article and offers his own “The role of humans in Google search” post on his blog. Matt says Google’s current methods include plenty of human input. The article was also covered by Eric Enge in “Human Input and Algorithmic Search” at Search Engine Watch Blog and by Barry Schwartz in “Google’s Human Touch” at Search Engine Land.

Our take? We still believe better results will come from implicit people-power — anonymous behavioral data used to distill search relevance that is not dependent on “explicit” user feedback. Think of John Battelle’s (his take on the NYT article here) “database of intentions” and the emerging concept of “the implicit web“.

A much richer user-generated dataset can be developed by avoiding the pitfalls of low participation rates (people won’t do the extra work) and of survey bias errors (partial or false information). Early implicit people-power successes will be applications related to site search and content discovery for specific websites. Later, as more data is collected, we will see improved relevance for web-wide index searches using implicit user feedback loops.

When you look at an ant farm or a beehive do you see chaos, or a highly organized system of segmented roles, responsibilities, and interests? Unless you are an entomologist, it might be difficult to understand what’s going on.

What about when you look at the activity on your website? Do you see “traffic” and “popular pages”? Or, do you see the behavior associated with your most important customer segments and the web pages most highly valued by them? There’s an obvious difference.

Frank Watson, at SEW, has recently pointed to two great articles in his “Is Segmentation Next Big Thing in Web Marketing” post. The first article, “Segmenting Search Intent”, by Rand Fishkin at seomoz.org, discusses how search intent (query keyword structure) translates into a useful customer segmentation perspective. The second article, “Visitor Segmentation Offers Target Marketing Insight” by Jennifer LeClaire at Newsfactor.com, provides additional insights on Web visitor segmentation.

Effective marketing normally relies on an understanding of the discrete customer groups that buy your products. Each group can be clustered based on a set of common characteristics or common search intents; the point of the articles above.

A natural byproduct of the Collarity platform is a rich detailed picture of the natural customer communities (segments) that live on a website. Collarity identifies specific “hotspots” of interest based on the measured associations of a web site’s visitors with content. These hotspots form the basis for organizing search and recommendation relevance by implicit communities of like-minded users and/or subject matter experts.

This segmented collective intelligence (Collarity Audience Analytics) can be used to better understand if you are taking good care of your most important customer groups on your website. It allows you to monitor their likes, dislikes, and how their tastes are changing over time.

We, of course, love discussions about the advancement of search technology and we were gratified to be highlighted this week in the “Top 17 Search Innovations Outside of Google” article and search innovation poll on Read/WriteWeb. The results, as of this post, are included to the right.

We think there are a couple of big things happening in search right now.

One, power is headed toward the people. For websites and web publishers, we think power is heading toward your audience. The foundation of search relevance is moving beyond power-link brokers and more toward end-user consumers on the edge. It’s easy to point to explicit feedback models like Digg, reddit, and social bookmarking sites. But we think there is an implicit customer feedback model that provides better results.

To better understand what’s happening, think about all our web analytic friends at the Emetrics Summit in San Francisco this week. They’re focused on trying to better understand online “customer engagement” through the all of the tiny pieces of of visitor behavior (searching, playing videos, stopping videos, scrolling lists) that tell a rich implicit story of user likes and dislikes. It’s all about actions on a page, and the pageview statistic going the way of the dinosaur.

Which brings us to the second big thing we see happening, search (most people think of Google) is blending with discovery (most people think of Amazon). Once user-engagement can be effectively measured and organized, people shouldn’t have to spend so much time searching for stuff — it should come to them. The same collective intelligence being gathered to refine search results can also be used to provide recommendations and advice — “visitors who searched/viewed this, also searched/viewed this”.

As search moves beyond links and pages, closer to people and actions, everything gets more relevant.

Business usually works best when decisions are based on data, instead of “gut”. The headwaters of truth, for many organizations, lives within their customer base. Want to know which way to vector product development? Ask your current customers. For websites, it’s important to understand the distinct communities of interest that live on your site, and the directions those segments are heading.

But how can you collect the right data? We’re all familiar with the “garbage in, garbage out” axiom and we know some sources of data provide the basis for better decisions than others.

We share a common philosophy with Greg Linden (founder of Findory and formerly a key developer with Amazon) regarding the advantage of implicit vs. explicit personalization and search mechanics. We both believe that a more accurate picture of a person’s interests and preferences can be developed from their implicit online behavior vs. any information they may decide to explicitly provide themselves.

For Collarity, this is especially important because we cluster this anonymous behavioral data, creating implicit communities around key subject areas of interest on a web site. We then provide community-driven site search and content discovery relevance from these clustered groups of “like-minded” visitors. In addition, we develop a rich source of business intelligence regarding the behavior of a site’s most important customer segments (the Pareto 20% that normally generates 80% of the revenue).

Greg succinctly outlined the primary problems with explicit personalization data way back in an April 2004 post on his blog:

When you rely on people to tell you what their interests are, they (1) usually won’t bother, (2) if they do bother, they often provide partial information or even lie, and (3) even if they bother, tell the truth, and provide complete information, they usually fail to update their information over time.

The primary advantages of a system that leverages existing implicit anonymous search and browsing behavior, like Collarity’s or Findory’s, is the following:

• You do not alienate your web visitors by requiring them to provide explicit feedback like tagging, rating, or bookmarking. Logically, it would be best to use a system that takes advantage of the implicit dialog already happening between a site and its audience, rather than burdening users with a new one.
• You move from a very small minority of people willing to tag/rate (often 1% or less) for the benefit of other users on a site, to 100% participation - providing a much richer and complete data set to distill relevance from.
• An implicit system is much more efficient and is capable of operating in realtime (very important for certain sites), because it doesn’t rely on users themselves to maintain explicit feedback.

Greg, in a post on his his blog last week, pointed to yet another paper being presented at the WWW 2007 conference this week titled, “Open User Profiles for Adaptive News Systems: Help or Harm?”, and again concludes that explicit mechanics can lead to less accurate results.

Greg also reminds us of a great quote from a Jason Fry recommendation engine article in the Wall Street Journal that summarizes implicit vs. explicit truisms most of us are all familiar with:

When it comes to describing us as customers and consumers, recommendation engines may do the job better than we would.

In other words, we lie — and never more effectively than when we’re lying to ourselves … I fancy myself a reader of contemporary literature and history books, but I mostly buy “Star Wars” novels and “Curious George” books for my kid.

So, whether you’re using customer feedback to refine site search results, generate content recommendations, or figure out “which way the wind is blowing” with your most important segments, we think user actions speak “truer” than words.

The main headline on the cover of Website Magazine this month is “Going Social”. WM has an in-depth article summarizing the online “wisdom of the crowd” phenomenon and they also include a “Going Social on Your Site” section that discusses how Web marketers can harness the power of social media.
So, we were flattered when Website Magazine chose to implement Collarity’s community-based site search and content guidance platform for use with their own site. Editor in Chief, Pete Prestipino, posted Website Magazine Integrates Collarity Relevance Engine on their blog last week summarizing their experience.

Website Magazine just integrated the Collarity Compass and some other additional features to enhance our in-site search capabilties. We integrated Collarity this morning on the WebsiteMagazine.com homepage, the search results page and late this afternoon on the blog and articles this afternoon. We’re still working out the kinks but excited about providing current and future site visitors with automated search and content suggestions, based on their own natural communities-of-interest.

Website Magazine is a well respected online resource that caters exclusively to the business of running a website.

Collarity was invited by the SIIA to participate in their Vendor-Buyer Technology Showcase session last week during the SIIA Content forum in San Francisco. John Blossom, founder of Shore Communications, provides a nice summary of the three companies (FAST, Nstein, and Collarity) who all had five minutes to present “the bottomline” of why the web publishers in the room should consider adopting their solutions.

All of the applications are designed to better connect web visitors with content after they arrive on a site. All of the presenters also did a good job of condensing presentations, that might normally take an hour to deliver, into easily digestible 5-minute summaries.

Many of the questions focused on the potential impact to existing legacy content systems, IT staff, and IT budgets. The session was an interesting comparison of technology and business models.

Post presentation feedback indicated that many attendees liked the idea of having their web audience drive search and recommendation relevance on their site. Most were interested in Collarity because we integrate easily with existing site index/content (we can also create the index if needed) via hosted web service (no software to install maintain), our installation is measured in days instead of months, and there are no up-up front/ongoing fees for publishers willing to share in the incremental advertising we deliver.

Collarity has been invited to provide a demonstration of our community-powered site search and content recommendation platform for web publishers on Tuesday, June 19th, at the O’Reilly Tools of Change for Publishing Conference. Collarity will be detailing several case studies of how our our service improves website content findability and monetization, leveraging the anonymous searching and browsing activity of previous visitors.

The conference is designed for anyone interested in understanding the latest trends and tools transforming the publishing industry. O’Reilly TOC runs from June 18th to the 20th, and will be held at the Fairmont Hotel, 170 South Market Street in San Jose.

Hope to see you there!

We’re always happy when our customers are happy. We’re even happier when they write about it, so we don’t have to.

Recently, all of the local Fox TV affiliates began using Collarity to provide search and content discovery for their web visitors. The Fox TV folks in Kansas City posted an announcement to let all of their audience know about the new upgraded service.

“I can’t say enough about this! You can finally find what you’re looking for. Navigation has always been our biggest challenge. And, while we’re still far from perfect. This is a big step in the right direction.”

We’ve been getting calls and emails asking what we think about Google’s recent announced shift toward more personalized search results and how their personalization compares to Collarity’s. The Google story has been widely reported by people like like Danny Sullivan of Search Engine Land and Kevin Newcomb at Search Engine Watch.

First, a quick clarifying point:

Collarity is a web service platform used by web publishers to provide search and content discovery services, based on the anonymous community-of-interest behavior of their audience. Personalization is an optional feature that publishers can choose to implement if they want to provide this service for their web visitors.

Second, rather than us trying to define the differences, we thought it would be easier to cite a recent post by one of our customers, who did just that. Lisa Barone writes in the Bruce Clay Blog about their recent experience using Collarity and how it compares to Google.

Lisa writes:

“The reason I like Collarity over a system like Google’s is that Collarity is non-intrusive. You don’t have to search within your personalized results. You decide how you want to search – from within your own search history (Personal), the history of your specific community (Community), or by pulling results from the entire Collarity network (Global). As long as users can select how they want to search and can easily opt out, personalized search has some benefit.”

“I think the best part is that you don’t have to register unless you want the personalized results. If you’re happy with just the community and the global results, you can stick to those and never sign in anywhere. That makes my tinfoil hat and me very happy.—Susan”