The Truth About Google’s Ereputation Algorithm

google shadow

google shadow

As the Web introduced a plethora of new social functionalities those past years, the amount of social data publicly available online exploded. In this metamorphosis, Google stayed the one-stop shop for anyone looking for information about anyone else. When your search query is a name, your search results typically show social profiles (Twitter, LinkedIn, Facebook, …), blogs, optionally video and/or photos results. In other words, as the Web got more social, Google remained the best information curator that could break a person’s life down to 10 links. That is why the best way to define Online Reputation is to summarize it to your first result page in Google.

For some individuals, it’s not always the good stuff that shows up in Google: Crooks, thieves, sociopaths, enrolled in a terrorist organization, people that choose to do wrong. Because of this, we’re lucky to have Google: The search engine spots negative content about them and brings it up to the front page. This way, you can avoid dealing with the wrong people.

All leads to think that Google is a good-doer, helping the helpless identify threats in a matter of minutes. The sad story behind this social justice Super Google story is that it’s nothing but a pulp fiction. Hear me out:

A lot (and I mean a lot) of people are experiencing a case of bad reputation with a site called truthaboutscientology.com. This site is on a quest to demystify the actual size and importance of the Scientology cult. To do so, the Website collects all public data available that contains names of people who attended any event organized by the cult that Tom Cruise dears. This means that if a friend invites you to attend this kind of event without informing you of the name of the organizer, your name would be listed, and through truthaboutscientology.com, this information would pop up in your first results in Google.

At first, I just thought that Google cannot be 100% accurate, and that making truthaboutscientology.com pop up in the first page could be a mistake. However, my opinion shifted a big way when I stumbled upon an old article written in 2002 by the owner of the site truthaboutscientology.com, called “Google refuses ad for critical site” (it’s a nofollow link, for whatever it’s worth today).

On this page, the author explains how in 2002 she started to promote her anti-scientology Website through Google Adwords. First she noticed that her ads were generating an awful lot of clicks. Second, after running for a day or two, Google stopped running her ads. Every time the anti-scientology Webmistress tried to launch a new Adwords campaign, Google would block it. After a long conversation between the now-victim of Google and mister “Don’t be Evil”, the Adwords rep concluded:

Thank you for advertising on Google. At this time, we are not running ads for sites that advocate against any individual, group, or organization. […] Google believes strongly in freedom of expression and therefore offers broad access to content across the web without censoring results. At the same time, we reserve the right to exercise editorial discretion when it comes to the advertising we accept on our site, as noted in our advertising terms and conditions. Please note that the decisions we make concerning advertising in no way affect the search results we deliver. We will continue to show search results for sites which advocate against individuals, groups or organizations.

Let me read this out in plain English for you: Google knows that truthaboutscientology.com hurts individuals, which is why Adwords won’t promote it, but the search engine has no problem hurting a person’s image through its search engine: “We will continue to show search results for sites which advocate against individuals, groups or organizations.”

There, you have it! This is why trying to sink Google’s first page of results with positive content will never take the stains away, simply because stains are an integrated part of Google’s favorite content. Once again, in the situation where the name queried belongs to a trustless indivdual, Google’s reputation component is a blessing. However, when Google search results show inaccurate negative content, there is a strong harm being done to the victims: Lives are literally being ruined because of Google’ algorithmic mistakes, or more globally because of Google’s approach to search results, trust and ereputation.

Truth or money?

So what is Google’s leitmotiv for raising the stakes so high on the quality of its ereputation algorithm? I like the idea that Google wants to crack the truth open into the wild, but I don’t see a reliable business model behind it 🙂 Technically, bringing up the negative truth does not generate clicks on sponsored results, unless…

Unless the victims of this algorithmic mess start to throw money into Adwords’ pipes to make positive and more accurate content re-conquer the first results on their names. The truth doesn’t pay the rent, but defamation and blackmailing does. Google’s reputation algorithm works like a Scarlet Letter system. In the Scarlet Letter, one woman was socially banned by her village because she had a child from an unknown man, and that woman was forced to wear the Scarlet Letter on her shirt to remind everyone that it was ok to spit on her. Religious beliefs motivated the members of the village to outcast the single mother. And what do we learn at the end of the story?

Google Algorithms, Twitter Crowdsources

Twitter crowd

Twitter crowd

Sarcasm on Twitter is common. Quite often, because of the short format of Twitter, sarcasm can be misleading. Smileys add a contextual meaning to short messages, and can prevent miscommunications in such cases as: “I’m gonna kill you!… :)” We don’t always think about it, mostly when we don’t think that our sarcasm could have negative repercussions.

Well, guys and gals, here is the story of the guy who got stroke by the most unlikely odds, and experienced his tweets growing into a community alert. Paul is the creator of Boagworld. Paul shares his story here, but I’ll give a shorter version below if you just want the gist of it.

Basically, on this day, Paul was leaving to visit his family, which did not really delight him:

Off to somerset to endure a ‘christmas meal’ with the extended family.

The weather that day was especially bad, which led Paul to write this:

This is turning into the journey from hell. Two roads closed because of floods. One road because of an accident.

And then, when he got to his family’s house and wasn’t happy about it, he wrote this via Brighkite:

Help me. Please help me. – http://bkite.com/02XsB

The geo-localization through Brighkite marked Paul in the middle of nowhere, so one of Paul’s followers on Twitter got scared. He thought Paul might have been a driving victim of the floods. The poor fellow called 911 right away (or whatever the number is in Great Britain), and the fire service dispatched two fire engines to attempt to find Paul.

I thought this was an amazing story. Not really for the two firetrucks being called out for nothing. Twitter is grabbing our attention because of the behavioral potential we can get out of it. With the power of three tweets, Paul activated a whole community alert.

On Twitter, everything you say is tracked, read and repeated. This shows how Twitter is successfully building a crowdsourcing intelligence that can detect threats and react to it. Also, I can’t help but compare Twitter’s community approach to detect threats and Google algorithmic approach to detect threats. The two companies are very holistic in their approach of the Web, and we are starting to see a few signs that indicate future competition in some areas between the two parties.

Thank you Paul for sharing this story.

HyveUp – Chris Law – Aggregate Knowledge (1)

aggregate knowledge

Last week, Facebook was announcing the Live Search integration into the popular social network search utility. The reactions on popular tech blogs were almost unanimous: Why?

Chris Law is the co-Founder and CEO of Aggregate Knowledge, a Web startup that specializes in serving recommendations on medium and high-traffic sites. Previously, Chris launched Tribe.net, an early social networking site. That social graphs are poor predictors for advertisers is the major lesson he learned from this startup.

Aggregate Knowledge doesn’t tap into social graphs to serve recommendations to its visitors. Instead, it uses a complex algorithm that analyzes two major dimensions of the visitor:

  1. Behavioral patterns
  2. Contextual patterns

The behavioral analysis is anonymous, so it doesn’t raise the same issues Beacon did. The contextual analysis is based on semantics. By mixing those two ingredients, Aggregate Knowledge serves up quality recommendations to its clients, who just have to insert a snippet of code in their site’s sidebar to get the service up and running. Aggregate Knowledge is a good example of a startup which development plans drift away from the social hype of the Web 2.0. In this post, Alex Iskold describes the challenges of building a recommendation engine. Since he is in the recommendation business, he has great analytical skills on that subject, and the post makes us understand how complex designing a behavioral/contextual recommendation engine can be.

In the same spirit of heavy data storing and crunching, Aggregate Knowledge’s approach is complex and powerful: Both Google and Microsoft invest heavily in behavioral targeting technologies, and semantics has been publicized as the new big trend many times (some competing recommendation engines focus exclusively on semantics).

However, Aggregate Knowledge isn’t a search tool, but a discovery tool. Discovery happens in a different context than search, a topic I will further expand on in the next post.

Read related items:
Comparing Discovery Tools (Whonu, Evri, Aggregate Knowledge)
3 Different Approaches to Automated Recommendation (Pandora, Strands, Aggregate Knowledge)
Recommendation Engines: Future for Retailers and Content Providers?

HyveUp – Todd Parsons – Buzzlogic

BuzzLogic

Yesterday, Todd Parsons, co-founder and Chief Product Officer over at Buzzlogic, came by my office for a short video interview. Buzzlogic recently launched the Conversation Targeting tool to help marketers efficiently spot and advertise on influential blogs.

The beauty of it is, it gives advertisers the ability to follow the Web users’ navigation around zones of high influence, and instantly place their ads accordingly. Buzzlogic has a high focus on blogs. Todd explains in the video how:

1. Blogs are search engine friendly
2. Potential buyers online use search engines to find opinions that they are most likely to find on blogging platforms.

Their technology measures blog content in terms of trust: Who is the best expert in a given field? Who do people listen to? Who’s going to significantly help me sell my product? Who can I trust for this job. Trust is determined with a set of algorithms that measure credibility and influence.

While they have developed the technology to spot those blogs where influence happens, Buzzlogic focuses now on developing tools to make it easier for blog publishers and companies to connect in a meaningful way.

Companies like Buzzlogic make me wonder about the future of media. These companies empower the long-tail Web by giving a chance to any great content to be heard and advertised. Great content naturally bubbles up to the top and C-listed bloggers can now simply focus on writing something great instead of sweating to get a link from a popular blog.

Those new algorithms offer a way for consumers to tackle conglomerates of media that were keeping a strong hold on mass information. A new media landscape is shaping up, one where information flows virally and every one is welcome to participate. The long-tail is a loose structure of isolated sources of information timidly interconnected.

Out of the chaos, thanks to powerful algorithms, new leaders emerge, new alliances form, and new ways to control the information arise. I’m glad I got to interview one of the main contributor to those changes.

For more info about Buzzlogic, visit their new video blog here.