Google Ranking Philosophy

Today, Google posted (finally) the first in what was promised (two months ago) to be a regular series of posts intended to open up more about aspects of internal operations. The original post was laden with disclaimers to the effect that certain truths could not be revealed due to the nature of their business. Since then, I've been waiting (not alone, I'm sure) to see just how much Google would be willing to share.

Google Announces Ad Planner

Today, Google announced Google Ad Planner, taking a direct shot at services offered by comScore and Nielsen. Differing from Google Trends for Websites, Ad Planner is aimed at media buyers looking to identify target sites for ad placements. In a market where Nielsen and comScore charge hefty sums to advertisers for access to their data while at the same time being loathed by publishers for the inaccuracies of their panel based measurement systems, Ad Planner is likely to gain acceptance fairly quickly.

There is also a level of integration with DoubleClick's MediaVisor product. The possibilities for integration between the DoubleClick and Ad Planner are endless and will give Ad Planner a distinct advantage over competing services. The service is currently in closed Beta right now but I'm hoping to get my hands on a login so I can test it out.

Stop Complaining about Google Trends for Websites

So, as I'm sure you're aware, late last week Google announced the release of Google Trends for Websites.If you haven't checked it out yet, or read one of the many reviews, it basically offers (approximate) site traffic volumes, top keywords, and related URLs for sites Google has a sufficient amount of data for.

Google joining the ranks of potentially (or known) inaccurate sources for traffic data (comScore, Alexa, I'm looking at you..) asides, a few folks got seriously up in arms over this release. SEOBook.com screamed "The Death of Privacy," and Lisa Barone (doing her best high school sophomore imitation) said "Not Cool, Google."

Why do people forget the contract we have with Google. We allow them to index and store our site content and use their free analytics tool without any pre-conditions on how they may use they data collected in the process. To be clear, I believe that Google wields way too much power over the Internet economy. But that doesn't matter.

If you don't want Google to have your data then stop using Google Analytics, uninstall the Google tool bar (which everybody loves to complain about), and block the GoogleBot using Robots.txt. If you want to continue to use their free services and benefit from the SEO-bubble we're all living in, then please shut up.

Google Webmaster Tools API

There are some things I don't like about Google, but then there are things like this that remind me of what an innovative company they are. No other company today embraces the internet community quite the way Google does and (most times, it seems) they truly strive to provide tools with real value.

Today, on the official google webmaster central blog, Google has announced the release of the Webmaster Tools API. While the initial function set is limited (outlined below), this is a nice way to provide access to webmaster data and promises to deliver greater functionality in the future.

Here's the list of currently supported features:

• Managing Sites
  • Retrieve a list of your sites in Webmaster Tools
  • Add your sites to Webmaster Tools
  • Verify your sites in Webmaster Tools
  • Remove your sites from Webmaster Tools
• Working with Sitemaps
  • Retrieve a list of your submitted Sitemaps
  • Add Sitemaps to Webmaster Tools
  • Remove Sitemaps from Webmaster Tools

An immediate use for the Webmaster Tools API (as functionality expands) would be an automated notification system for crawl errors, Sitemap errors, etc. I'm sure I can think of a few more.

What creative uses do you think a fully functional Webmaster Tools API could be used for?

Google: More Freshness in Every Query

Yesterday Google announced updates to their query recommendation algorithm that will process newer information in a more timely manner.

I tried a few searches for items that have recently been in the news and the results were good - for example, a search for "Iowa" recommended "Iowa Flooding". The same search, however, also recommended "Iowa DOT." I suppose the state department of transportation would be in high demand given the recent flooding, but seeing that recommendation is occurs to me that what's missing from these suggestions is context.

It would be helpful to the searcher is Google could present some bits of information indicating why a given search term is being suggested. Here's another example - I searched for "Father's Day" and among the suggested queries was "Father's Day, 2007." As a searcher, it's unclear to me why I should be interested in Father's Day from last year. Providing some of the deterministic data along with these recommendations would help me understand the relevance of this recommendation to my intentions. After all, relevance is what Google is all about.

Google + Yahoo

TechCrunch is reporting that Google and Yahoo are announcing (right about now) a partnership that will most likely involve outsourcing Yahoo's search business to Google. Given the amount of dominance Google already has in search this will certainly raise market concerns as such a deal could push Google's share of the search market to near 90%.

I've been fairly vocal in the past in my opinion that Google has way too much power over the internet economy at present and this certainly doesn't help matter. For me, I'd much rather have seen the Microsoft deal go through so somebody, somewhere, could introduce real competition into the market. Online search is in dire need of innovation at the moment and giving Google a bigger slice of the pie will do nothing to foster that innovation, regardless of how creative the company has been in the past.

 

Update: For those of you interested in the gritty details, TechCrunch also has the the Yahoo SEC Filing which outlines some specifics of the deal.

Any other thoughts on this out there?

Whither Google Transparency?

A few weeks back Udi Manber, the VP of Engineering for Search Quality, made a much publicized post on the Official Google Blog. In the post, he promised an increased level of transparency from the search giant and told us to watch that space for future openings of the kimono.

Well, since that first bit of bait, here's what we've gotten -

Google Search Quality

Late yesterday, Udi Manber, VP of Engineering for Search Quality over at Google offered up a post on the Official Google Blog titled - Introduction to Google Search Quality. What drew me in was this quote -

This blog post is part of a renewed effort to open up a bit more than we have in the past.

Having high hopes I read on, hoping for some insights into the (largely) black box that is search relevancy. Unfortunately, I was disappointed. The post dedicated a fair amount of time to detailing just how hard the job of the Search Quality team is, and how every second they have to handle hundreds of queries, parsing billions of documents to pluck out and order the relative few. And how they have just milliseconds to do so. Some time was also spent on the structure of the group, internationalization efforts, and some past history.

That's all great, but it feels like a setup to me. Frankly, this post did nothing to "open up a bit more" and even reiterated the point of why they need to be so secretive -

We are, to be honest, quite secretive about what we do. There are two reasons for it: competition and abuse. Competition is pretty straightforward. No company wants to share its secret recipes with its competitors. As for abuse, if we make our ranking formulas too accessible, we make it easier for people to game the system. Security by obscurity is never the strongest measure, and we do not rely on it exclusively, but it does prevent a lot of abuse.

So in one blog-breath Google tells us their opening the door a crack while at the same time applying a few more dead bolts. It smells of PR to me and not a genuine desire for transparency.

I was also surprised to see this item receive fairly sparse coverage around the blogs and news sites. I don't go crazy with my RSS subscriptions like lots of folks do, but I've got 40 or so relevant ones in Google Reader and I've only seen this bit from SearchEngineLand - and that's nothing more than a recap  without any commentary.

Perhaps I'm being too quick to judge. The post is, after all, labeled as the first in a series. So far, nothing has been revealed, but I'll be watching that spot in Google Reader closely over the next few weeks.

Related Posts: The Expanding Web

Google Friend Connect

Yesterday Google announced a private beta of FriendConnect - a standards based (OpenID, oAuth, and OpenSocial) collection of widgets that can bring social networking type functionality to any site or blog. The announcement has garnered its fair share of coverage (Mashable, Yahoo!, SearchEngineLand, Outside the Lines) so instead of simply rehashing what's already been written about FriendConnect, I'd like to look a little deeper.

With the CPC contextual text link model, Google opened up online advertising to the masses. With AdWords, even the smallest web site could bid on keywords and arbitrage traffic - effectively eliminating all barriers to entry. With FriendConnect, we will see the same effect. Any web site, no matter how small, will be able to add social networking components. Initial availability will include user registration, invitations, member galleries, message posting, and reviews, as well as OpenSocial applications.

This is again Google's long-tail approach to the web. Commodotize a feature set once controlled by a small number of vendors, and in so doing reach the vast potential of 100+ million web sites out there. With FriendConnect, Facebook and MySpace lose their prominence as social media platforms and the web itself becomes the platform. Consider how the game changes when your OpenID follows you from site to site and you know which friends are active on which sites. Combine this with the data portability movement and suddenly the web begins to resemble, well, life. Walk into any room, see whos there, recognize your pals and, wherever you go, you're the same person.

Google Toolbar PageRank Gone

On the heels of a highly isolated StumbleUpon disappearance, my string of internet anomalies continues. The PageRank has gone missing from the google toolbar. As of this morning, for every site I visit, my google toolbar in both IE and FIrefox has lost the little green progress bar. Upon rollover, I get the following message "No PageRank information available."

I checked a few other computers in the office and we're all seeing the same thing. Anybody else out there seeing this?